A method for adjusting roasting process parameters based on low-temperature synergistic extraction of tea pigments

Abstract

This invention relates to the field of tea testing technology, and more particularly to a method for adjusting roasting process parameters based on low-temperature synergistic extraction of tea pigments. The method includes: determining whether the tea raw material is a first-stage light-roasted tea based on images of the dry tea leaves, and classifying non-first-stage light-roasted tea leaves; determining the roasting grade of the heavily roasted tea based on leaf images; determining the tea pigment deviation ratio by combining the proportion of tea pigment components in the second-stage heavily roasted tea; adjusting the roasting process parameters based on the difference between the tea pigment deviation ratio and the corresponding preset tea pigment deviation ratio; determining the optimized deviation ratio, judging whether the adjusted roasting process parameters meet the preset standard, and determining the cause of the formation of the third-stage heavily roasted tea; and determining the tea pigment composition of the third-stage heavily roasted tea based on the tea infusion images of the acquired third-stage heavily roasted tea and standard tea infusion images. This invention can improve the accuracy of visual inspection of tea.

Description

Technical Field

[0001] This invention relates to the field of tea detection technology, and in particular to a low-temperature synergistic extraction method for tea pigments. Background Technology

[0002] In the field of deep tea processing, low-temperature extraction technology has become the mainstream process for extracting tea pigments because it can preserve the active ingredients and natural properties of tea pigments to the greatest extent. The visual inspection method for tea based on this technology was developed to meet the needs of rapid screening of raw material quality, real-time monitoring of extraction efficiency, and precise control of finished product quality in the tea pigment production process. Traditional tea quality inspection relies heavily on sensory evaluation, which is highly subjective and difficult to quantify tea pigment-related indicators. It cannot meet the strict requirements of low-temperature extraction process on the content, stability, and composition ratio of tea pigments in raw tea leaves. Therefore, it is necessary to establish an objective and quantifiable evaluation system through visual inspection technology to link the appearance characteristics of tea with the actual extraction potential of tea pigments.

[0003] Chinese Patent Application Publication No. CN120472234A discloses a method for determining the fermentation degree of tea using image recognition. This invention includes capturing high-definition images of tea leaves at different stages of processing using a high-definition camera, preprocessing the images using noise reduction, contrast enhancement, and color correction techniques, converting the images to a suitable color space for extracting color features, and analyzing the dominant hue and color distribution on the tea surface. This invention relates to the field of tea detection technology. This method enables non-destructive, real-time, and accurate monitoring of tea fermentation degree. By capturing tea images and analyzing their color, shape, and texture visual features, combined with machine learning algorithms, the fermentation state of the tea can be automatically determined, thereby improving the automation level and production efficiency of the tea processing process.

[0004] However, the above method has the following problems: simply observing the appearance of tea leaves is not accurate enough for judging the degree of fermentation and the tea pigments themselves. Summary of the Invention

[0005] Therefore, the present invention provides a low-temperature synergistic extraction method for tea pigments, which overcomes the problem that the existing technology is not accurate enough in judging the degree of tea fermentation and the tea pigments themselves by simply observing the appearance image of tea leaves.

[0006] To achieve the above objectives, the present invention provides a method for low-temperature synergistic extraction of tea pigments, comprising: Based on the acquired dry tea image, it is determined whether the tea raw material is a first light roast tea. Based on the acquired tea soup image of non-first light roast tea, the non-first light roast tea is classified. Collect leaf images of heavily roasted tea leaves, and determine the roasting grade of the heavily roasted tea leaves based on the leaf images; Standard tea pigments were extracted from the first-grade dark roasted tea leaves to determine the proportion of standard tea pigment components and standard tea infusion images. The proportion of tea pigment components in the second-grade dark roasted tea leaves was then used to determine the tea pigment deviation proportion. Adjust the baking process parameters based on the ratio difference between the tea pigment deviation ratio and the corresponding preset tea pigment deviation ratio; The optimized proportion of tea pigment components in the heavily roasted tea produced after adjusting the roasting process parameters is determined by combining the proportion of the standard tea pigment components. Based on the optimized deviation proportion, it is determined whether the adjusted roasting process parameters meet the preset standard and the cause of the formation of the third roasting grade of heavily roasted tea is determined. The tea pigment composition of the third-grade dark roasted tea was determined based on the tea liquor image obtained from the third-grade dark roasted tea and the standard tea liquor image. The tea pigments include theaflavins, thearubigins, and theabrownins, and the tea pigment deviation ratios include the theaflavin deviation ratio, thearubigin deviation ratio, and theabrownin deviation ratio.

[0007] Furthermore, the process of determining whether the tea raw material is a first lightly roasted tea based on the obtained dry tea image includes, The tea leaves are laid flat on a white background to obtain an initial image of the dry tea. The initial dry tea image is cropped to obtain the dry tea image, and the dry tea image is denoised and converted to the HSV color model. The proportion of green pixel areas in the dry tea image is calculated and recorded as the light roasting ratio. Based on the light roasting ratio, it is determined whether the tea raw material is a first light roasted tea.

[0008] Furthermore, the process of classifying the non-first light roast tea leaves includes, Acquire images of tea infusions from non-first-light-roasted tea leaves and convert them to an HSV color model; The brightness value is calculated based on the tea soup image; The brightness value is compared with the first preset brightness value and the second preset brightness value respectively, and the non-first light roast tea is classified according to the comparison result; The classification of non-first light roast tea includes second light roast tea, medium roast tea, and heavy roast tea.

[0009] Furthermore, the process of determining the baking grade based on the leaf base image includes, Remove the fully expanded tea leaves after brewing and lay them flat on a white background to obtain the initial leaf image; The initial leaf base image is cropped to obtain the leaf base image, and the leaf base image is converted to the HSV color model; The proportion of green pixel areas in the leaf base image is calculated and recorded as the leaf base ratio. The baking grade is determined based on the leaf base ratio. The leaf base image is an image that only retains the base of the tea leaves.

[0010] Furthermore, the baking grade is determined based on the leaf-to-root ratio, wherein, If the leaf-to-root ratio is less than the first preset leaf-to-root ratio, then the baking grade is determined to be the first baking grade; If the leaf-to-root ratio is greater than or equal to the first preset leaf-to-root ratio and less than the second preset leaf-to-root ratio, then the baking grade is determined to be the second baking grade. If the leaf-to-root ratio is greater than or equal to the second preset leaf-to-root ratio, then the baking grade is determined to be the third baking grade; The first preset leaf-to-bottom ratio and the second preset leaf-to-bottom ratio are negatively correlated with the baking time, and the first preset leaf-to-bottom ratio is less than the second preset leaf-to-bottom ratio.

[0011] Furthermore, the process of determining the proportion of standard components of tea pigments includes, Under the condition that the roasting grade is determined to be the first roasting grade, the tea pigment in the corresponding heavily roasted tea leaves is extracted by ultrasonic low temperature method and recorded as the standard tea pigment; The proportions of standard tea pigment components were obtained by analyzing the standard tea pigments using spectrophotometry. The baking process parameters include baking temperature, baking time, and oxygen content.

[0012] Furthermore, the process of determining the tea pigment deviation ratio includes, Under the condition that the roasting grade is determined to be the second roasting grade, the tea pigments in the corresponding heavily roasted tea leaves are extracted by ultrasonic low temperature method. The proportions of tea pigment components in the corresponding heavily roasted tea leaves were obtained by analyzing the tea pigments in the tea pigments using spectrophotometry. The deviation ratio of each tea pigment is determined by combining the proportion of the tea pigment components and the proportion of the standard tea pigment components.

[0013] Furthermore, the process of adjusting the baking process parameters includes, Calculate several ratio differences between the deviation ratio of each tea pigment and the corresponding preset deviation ratio of tea pigment; The baking parameters are adjusted based on the comparison results between the stated ratio differences and the preset differences.

[0014] Further, based on the optimized deviation ratio, determining whether the adjusted baking process parameters meet the preset standard includes, If the optimization deviation ratio is less than the preset optimization deviation ratio, it is determined that the adjustment of the roasting process parameters meets the preset standard and that the reason for the formation of the third roasting grade of heavy roasted tea is the control deviation of the roasting process parameters. If the optimization deviation ratio is greater than or equal to the preset optimization deviation ratio, it is determined that the adjusted roasting process parameters do not meet the preset standard and that the cause of the third roasting grade of heavy roasted tea is a problem with the tea raw materials.

[0015] Furthermore, the process of determining the tea pigment composition of the third-grade dark roasted tea based on the tea infusion image of the acquired third-grade dark roasted tea and the standard tea infusion image includes, Tea infusion parameters are obtained based on the tea infusion image of the third roasting grade of dark roasted tea leaves. Obtain standard tea infusion parameters based on the standard tea infusion image; Calculate the similarity value between the tea infusion parameters and the standard tea infusion parameters, and determine the tea pigment composition of the third roasting grade of dark roasted tea based on the similarity value.

[0016] Compared with existing technologies, the advantages of this invention are as follows: This invention obtains an initial dry tea image by laying the tea raw material flat on a white background, cropping the initial dry tea image to obtain a final dry tea image, denoising the dry tea image, converting it to an HSV color model, calculating the proportion of green pixels in the dry tea image as the light roast ratio, and determining whether the tea raw material is a first-light roast tea based on the light roast ratio. This eliminates the need for complex sample pretreatment, significantly reducing labor and equipment costs, while avoiding the destructive consumption of tea samples by traditional chemical detection methods. It achieves non-destructive testing of tea raw materials, making it particularly suitable for rapid initial screening of batches of tea raw materials on production lines, effectively improving the reliability of the judgment criteria. The cropping operation can remove irrelevant background from the image edges. Focusing on the main tea leaf area reduces interference from irrelevant information. Noise reduction eliminates noise caused by light fluctuations and equipment noise during image acquisition, ensuring the authenticity of image color information. Converting the image to the HSV color model, compared to the RGB model, better aligns with human color perception, decomposing color into three independent dimensions: hue (H), saturation (S), and lightness (V). The hue dimension more accurately distinguishes the green tint of the tea leaves, laying a high-quality data foundation for subsequent calculations of the green pixel ratio and avoiding judgment bias caused by inappropriate color model selection. Compared to the dark brown / burnt black hues of heavily roasted tea, lightly roasted tea retains more of the original pigments from the fresh leaves, resulting in a certain proportion of green areas in the dry tea. Quantifying the green pixel ratio as a judgment indicator not only effectively avoids the risk of misjudgment due to differences in human experience but also standardizes the judgment results by setting a clear threshold for the light roasting ratio, ensuring consistency in detection results and improving the accuracy of visual inspection of tea.

[0017] Furthermore, this invention categorizes non-first-light-roasted tea leaves, obtains images of the tea infusion from these leaves, converts them to an HSV color model, calculates brightness values ​​based on the images, and compares these values ​​with a first preset brightness value and a second preset brightness value. Based on the comparison results, the non-first-light-roasted tea leaves are categorized. These non-first-light-roasted tea leaves encompass different roasting levels, such as medium and heavy roasting, which are difficult to accurately distinguish based solely on the appearance of the dry tea. The dry tea color of some medium-roasted teas may be similar to that of lightly and heavily roasted teas, easily leading to confusion. The color of the tea infusion is a direct reflection of the dissolved pigments within the tea leaves. The deeper the roasting degree, the higher the proportion of theaflavins and thearubigins in the tea leaves that polymerize into theabrownins, resulting in a darker tea infusion color and a lower brightness value. By introducing tea infusion image analysis, a two-dimensional judgment criterion is formed, effectively avoiding misjudgment caused by similar colors in a single dry tea image, significantly improving the accuracy of classifying non-first-level light roasted teas. Converting tea infusion images to the HSV color model can decompose color information into three independent dimensions: hue, saturation, and lightness. Among them, the lightness value directly reflects the brightness of the tea infusion and is not affected by hue and saturation, which can more accurately quantify the color characteristics of the tea infusion and further improve the accuracy of visual inspection of tea.

[0018] Furthermore, this invention determines the roasting grade based on the leaf bottom image. After brewing, the fully expanded tea leaves are laid flat on a white background to obtain an initial leaf bottom image. This initial image is then cropped to obtain a final leaf bottom image. The final leaf bottom image is converted to an HSV color model, and the proportion of green pixels in the image is calculated as the leaf bottom ratio. The roasting grade is determined based on this ratio. This process uses fully expanded leaves after brewing as the inspection object, filling the information gap in the inspection of dry tea and tea soup. Lightly roasted tea leaves retain more chlorophyll from the fresh leaf stage due to the weaker effect of high temperature, exhibiting a distinct green hue. As the roasting degree increases, the chlorophyll gradually decomposes, the proportion of green in the leaf bottom decreases, and it turns yellowish-brown or brown. By focusing on the leaf bottom image, the judgment basis can be supplemented from the dimension of "internal organizational characteristics of tea leaves", effectively avoiding misjudgment caused by external factors in the single dimension inspection of dry tea or tea soup, making the roasting grade judgment more in line with the true quality state of tea leaves. The cropping operation can remove irrelevant areas of the white background in the leaf bottom image, focus on the fully unfolded tea leaf body, avoid the interference of background noise on color judgment, and further improve the accuracy of visual inspection of tea leaves.

[0019] Furthermore, this invention compares the differences in tea pigments between first-grade and second-grade heavy-roasted tea leaves. It adjusts the roasting parameters of insufficiently roasted second-grade heavy-roasted tea leaves to obtain first-grade heavy-roasted tea leaves that meet the standards. This also allows for accurate quality control of the raw materials used for subsequent tea pigment extraction. Additionally, based on the analysis of the adjusted second-grade heavy-roasted tea leaves, it determines whether sufficient roasting has been achieved, thus identifying the reasons for the formation of third-grade heavy-roasted tea leaves. If the analysis shows that the adjusted second-grade heavy-roasted tea leaves have achieved sufficient roasting, it indicates that adjusting the roasting process parameters can control the roasting process. The deviation in roasting, specifically the formation of the third-grade heavy roasted tea, is due to deviations in the control of roasting process parameters. If the adjustment results of the second-grade heavy roasted tea, after adjustment, do not achieve sufficient roasting, it indicates that the initial substances for forming tea pigments in the raw materials of the heavy roasted tea are insufficient. In other words, the formation of the third-grade heavy roasted tea is due to a problem with the tea raw materials. Therefore, for the third-grade tea whose formation is determined to be due to a problem with the tea raw materials, its tea soup image is extracted and compared with the standard tea soup image of the same tea. Based on the similarity value, the tea pigment composition of the third-grade heavy roasted tea is determined, further improving the accuracy of visual inspection of tea. Attached Figure Description

[0020] Figure 1 This is a schematic flowchart of the low-temperature synergistic extraction method for tea pigments according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the process for determining the baking grade according to an embodiment of the present invention; Figure 3 This is a logic diagram for determining the baking grade in an embodiment of the present invention; Figure 4 This is a logic diagram for determining whether adjusting baking process parameters conforms to preset standards in an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0022] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0023] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0024] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0025] Please see Figure 1 The diagram shown is a schematic flowchart of a low-temperature synergistic extraction method for tea pigments according to an embodiment of the present invention. The present invention provides a low-temperature synergistic extraction method for tea pigments, comprising: Step S1: Determine whether the tea raw material is a first light roasted tea based on the dry tea image obtained; classify the non-first light roasted tea based on the tea soup image obtained from the infusion of non-first light roasted tea. Step S2: Collect leaf images of heavily roasted tea leaves and determine the roasting grade of the heavily roasted tea leaves based on the leaf images; Step S3: Extract the standard tea pigments from the first roasting grade of dark roasted tea leaves to determine the proportion of standard tea pigment components and the standard tea infusion image, and combine the proportion of tea pigment components from the second roasting grade of dark roasted tea leaves to determine the tea pigment deviation proportion. Step S4: Adjust the roasting process parameters based on the ratio difference between the tea pigment deviation ratio and the corresponding preset tea pigment deviation ratio; Step S5: Extract the optimized proportion of tea pigment components from the re-roasted tea generated after adjusting the roasting process parameters, and determine the optimization deviation proportion by combining it with the standard proportion of tea pigment components. Based on the optimization deviation proportion, determine whether the adjusted roasting process parameters meet the preset standard and determine the cause of the formation of the third roasting grade of re-roasted tea. Step S6: Determine the tea pigment composition of the third-grade dark roasted tea based on the obtained tea liquor image of the third-grade dark roasted tea and the standard tea liquor image. Among them, tea pigments include theaflavins, thearubigins and theabrownins, and tea pigment deviation ratios include theaflavin deviation ratios, thearubigin deviation ratios and theabrownin deviation ratios.

[0026] Understandably, when laying out tea leaves, stacking, overlapping, or localized dense areas should be avoided. A high-purity matte white background can be chosen. When cropping the initial dry tea image, the tea leaf distribution boundaries must be strictly followed, removing only the edges of the background and clutter from the shooting environment. After converting the dry tea image to the HSV color model, the hue (H), saturation (S), and lightness (V) range thresholds for "green pixels" must be clearly defined. Verification using multiple known first-light-roasted tea leaf samples is necessary to avoid misclassifying yellowish-brown pixels as green due to excessively wide thresholds, or missing light green pixels due to excessively narrow thresholds.

[0027] Understandably, when calculating the proportion of green pixel areas, it is necessary to count the total number of all pixels in the dry tea image that meet the green threshold, and divide it by the total number of pixels in the main tea area. If a single sample image has local color anomalies (such as individual tea leaves with scorched edges), it is possible to collect 3-5 parallel dry tea images and take the average of the green pixel proportions as the final light roasting ratio to reduce the impact of random errors in a single image.

[0028] Specifically, in step S1, the process of determining whether the tea raw material is a first lightly roasted tea based on the acquired dry tea image includes, Step S101: Lay the tea leaves flat on a white background to obtain an initial dry tea image; Step S102: Crop the initial dry tea image to obtain a dry tea image, perform noise reduction processing on the dry tea image, and convert it to the HSV color model; Step S103: Calculate the proportion of green pixel area in the dry tea image and record it as the light roasting ratio. Based on the light roasting ratio, determine whether the tea raw material is the first light roasted tea.

[0029] The light roast ratio is compared with the preset light roast ratio. If the light roast ratio is less than or equal to the preset light roast ratio, the tea raw material is determined to be the first light roast tea. If the light roast ratio is greater than the preset light roast ratio, the tea raw material is determined to be a non-first light roast tea. In one specific embodiment, a preset light roasting ratio of 20% is set. If the light roasting ratio of 14% is less than the preset light roasting ratio, the tea raw material is determined to be the first light roasting tea. If the light roast ratio is 28%, which is greater than the preset light roast ratio, then the tea raw material is determined to be a non-first light roast tea. It is understandable that as the degree of tea roasting increases, the proportion of green pixel areas in the dry tea image decreases. The preset light roasting ratio is negatively correlated with the tea roasting time. Preferably, the preset light roasting ratio is in the range of 20% to 30%.

[0030] Specifically, this invention obtains an initial dry tea image by laying tea leaves flat on a white background, cropping the initial dry tea image to obtain a final dry tea image, denoising the dry tea image, converting it to an HSV color model, calculating the proportion of green pixels in the dry tea image as the light roast ratio, and determining whether the tea leaves are first-light roasted based on the light roast ratio. This method eliminates the need for complex sample pretreatment, significantly reducing labor and equipment costs, while avoiding the destructive consumption of tea samples by traditional chemical testing methods. It achieves non-destructive testing of tea leaves, making it particularly suitable for rapid initial screening of batches of tea leaves on production lines, effectively improving the reliability of the judgment criteria. The cropping operation removes irrelevant background from the image edges, focusing on the tea leaves themselves. The text discusses various methods for determining the color accuracy of tea leaves. It mentions denoising to eliminate noise caused by light fluctuations and equipment noise during image acquisition, ensuring the authenticity of color information. Conversion to the HSV color model, compared to the RGB model, better aligns with human color perception. This model decomposes color into three independent dimensions: hue (H), saturation (S), and lightness (V). The hue dimension more accurately distinguishes the green tint of tea leaves, providing a high-quality data foundation for calculating the proportion of green pixels and avoiding biases caused by inappropriate color model selection. Compared to the dark brown / burnt black hues of heavily roasted tea, lightly roasted tea retains more of the original pigments from fresh leaves, resulting in a certain proportion of green areas in the dry tea. Quantifying the proportion of green pixels as a criterion not only effectively avoids the risk of misjudgment due to differences in human experience but also standardizes the results by setting a clear threshold for the light roasting ratio, ensuring consistency and improving the accuracy of visual inspection of tea.

[0031] Specifically, in step S1, the process of classifying non-first light roast tea leaves includes, Step S104: Obtain the image of tea soup brewed from non-first light roasted tea leaves and convert it to the HSV color model; Step S105: Calculate the brightness value based on the tea soup image; Step S106: Compare the brightness value with the first preset brightness value and the second preset brightness value respectively, and classify the non-first light roast tea according to the comparison result; If the brightness value is less than the first preset brightness value, then the tea that is not lightly roasted is determined to be heavily roasted tea. If the brightness value is greater than or equal to the first preset brightness value and less than the second preset brightness value, then the tea that is not lightly roasted is determined to be medium roasted tea. If the brightness value is greater than or equal to the second preset brightness value, then the tea leaves that are not the first lightly roasted tea leaves are determined to be the second lightly roasted tea leaves.

[0032] The classification of non-first-light-roasted tea includes second-light-roasted tea, medium-roasted tea, and heavy-roasted tea.

[0033] It is understandable that the first preset brightness value is less than the second preset brightness value. This is because the longer the tea is roasted, the more pigments accumulate in the tea soup, the weaker its ability to reflect light, and the lower the brightness value. The first preset brightness value and the second preset brightness value are negatively correlated with the tea roasting time. Preferably, the first preset brightness value is in the range of 25 to 40, and the second preset brightness value is in the range of 80 to 110.

[0034] In one specific embodiment, a first preset brightness value is set to 40, and a second preset brightness value is set to 110. If the brightness value is 38, which is less than the first preset brightness value, then the tea that is not lightly roasted is determined to be heavily roasted tea. If the brightness value is 63, which is greater than the first preset brightness value and less than the second preset brightness value, then the tea that is not lightly roasted is determined to be medium roasted tea. If the brightness value is 124, which is greater than the second preset brightness value, then the tea leaves that are not the first lightly roasted tea leaves are determined to be the second lightly roasted tea leaves.

[0035] Specifically, this invention categorizes non-lightly roasted tea leaves, obtains images of the tea infusion from these leaves, converts them to an HSV color model, calculates brightness values ​​based on the images, and compares these values ​​with a first preset brightness value and a second preset brightness value. Based on the comparison results, the non-lightly roasted tea leaves are categorized. These non-lightly roasted tea leaves encompass different roasting levels, such as medium and heavy roasting, which are difficult to accurately distinguish based solely on the appearance of the dry tea. Furthermore, the color of some medium-roasted tea leaves may resemble that of lightly and heavily roasted tea leaves, leading to confusion. The color of the tea infusion is a direct reflection of the dissolved pigments within the tea leaves. The deeper the roasting degree, the higher the proportion of theaflavins and thearubigins in the tea leaves that polymerize into theabrownins, resulting in a darker tea infusion color and a lower brightness value. By introducing tea infusion image analysis, a two-dimensional judgment criterion is formed, effectively avoiding misjudgment caused by similar colors in a single dry tea image, significantly improving the accuracy of classifying non-first-level light roasted teas. Converting tea infusion images to the HSV color model can decompose color information into three independent dimensions: hue, saturation, and lightness. Among them, the lightness value directly reflects the brightness of the tea infusion and is not affected by hue and saturation, which can more accurately quantify the color characteristics of the tea infusion and further improve the accuracy of visual inspection of tea.

[0036] Please see Figure 2 As shown, this is a flowchart illustrating the process of determining the baking grade according to an embodiment of the present invention. In step S2, the process of determining the baking grade based on the leaf bottom image includes: Step S201: Take out the fully expanded, heavily roasted tea leaves after brewing and lay them flat on a white background to obtain an initial leaf image; Step S202: Crop the initial leaf base image to obtain the leaf base image, and convert the leaf base image to the HSV color model; Step S203: Calculate the proportion of green pixel area in the leaf bottom image and record it as the leaf bottom ratio. Determine the baking grade based on the leaf bottom ratio. Among them, the leaf base image is an image that only retains the bottom of the tea leaves.

[0037] Please see Figure 3 As shown, this is a logic diagram for determining the baking grade according to an embodiment of the present invention. In step S203, the baking grade is determined based on the leaf-to-bottom ratio, wherein... If the leaf-to-root ratio is less than the first preset leaf-to-root ratio, the baking grade is determined to be the first baking grade; If the leaf-to-root ratio is greater than or equal to the first preset leaf-to-root ratio and less than the second preset leaf-to-root ratio, then the baking grade is determined to be the second baking grade. If the leaf-to-root ratio is greater than or equal to the second preset leaf-to-root ratio, the baking grade is determined to be the third baking grade. Among them, the first preset leaf bottom ratio and the second preset leaf bottom ratio are negatively correlated with the baking time, and the first preset leaf bottom ratio is less than the second preset leaf bottom ratio.

[0038] It is understandable that the longer the baking time, the smaller the proportion of green pixel area in the leaf base. Therefore, the first preset leaf base ratio and the second preset leaf base ratio are negatively correlated with the baking time. Preferably, the value range of the first preset leaf base ratio is 3% to 5%, and the value range of the second preset leaf base ratio is 6% to 8%.

[0039] In one specific embodiment, a first preset leaf-to-bottom ratio of 3% and a second preset leaf-to-bottom ratio of 6% are set. If the leaf-to-bottom ratio of 2% is less than the first preset leaf-to-bottom ratio, the baking grade is determined to be the first baking grade. If the leaf-to-root ratio is 5%, which is greater than the first preset leaf-to-root ratio but less than the second preset leaf-to-root ratio, then the baking grade is determined to be the second baking grade. If the leaf-to-root ratio is 8%, which is greater than the second preset leaf-to-root ratio, then the baking grade is determined to be the third baking grade.

[0040] Specifically, in step S3, the process of determining the proportion of standard components of tea pigments includes, Step S301: Under the condition that the roasting grade is determined to be the first roasting grade, the tea pigment in the corresponding heavy roasted tea leaves is extracted by ultrasonic low temperature method and recorded as the standard tea pigment. Step S302: Analyze the standard tea pigments using spectrophotometry to obtain the proportion of standard tea pigment components; The baking process parameters include baking temperature, baking time, and oxygen content.

[0041] It is understandable that for those skilled in the art, using spectrophotometry to analyze standard tea pigments to obtain the proportion of standard tea pigment components is existing technology, and will not be elaborated here.

[0042] Specifically, this invention determines the roasting grade based on leaf bottom images. After brewing, fully expanded tea leaves are laid flat on a white background to obtain an initial leaf bottom image. This initial image is then cropped to obtain a final leaf bottom image. The final leaf bottom image is converted to an HSV color model, and the proportion of green pixels in the image is calculated as the leaf bottom ratio. The roasting grade is determined based on this ratio. This process uses fully expanded leaves after brewing as the inspection object, filling the information gap in dry tea and tea soup inspection. Lightly roasted tea leaves retain more chlorophyll from their fresh leaf stage due to the weaker effect of high temperatures, exhibiting a distinct green hue. As the roasting degree increases, the chlorophyll gradually decomposes, the proportion of green in the leaf bottom decreases, and it turns yellowish-brown or brown. By focusing on the leaf bottom image, the judgment basis can be supplemented from the dimension of "internal organizational characteristics of tea leaves", effectively avoiding misjudgment caused by external factors in the single dimension inspection of dry tea or tea soup, making the roasting grade judgment more in line with the true quality state of tea leaves. The cropping operation can remove irrelevant areas of the white background in the leaf bottom image, focus on the fully unfolded tea leaf body, avoid the interference of background noise on color judgment, and further improve the accuracy of visual inspection of tea leaves.

[0043] Specifically, in step S3, the process of determining the tea pigment deviation ratio includes, Step S303: Under the condition that the roasting grade is determined to be the second roasting grade, the tea pigments in the corresponding heavy roasted tea leaves are extracted by ultrasonic low temperature method. Step S304: Use spectrophotometry to analyze the tea pigments in the corresponding heavily roasted tea leaves to obtain the proportion of tea pigment components; Step S305: Determine the deviation ratio of each tea pigment by combining the proportion of tea pigment components and the proportion of standard tea pigment components.

[0044] It is understandable that the theaflavins deviation ratio is the absolute value of the difference between the theaflavins component ratio and the standard theaflavins component ratio, the thearubigins deviation ratio is the absolute value of the difference between the thearubigins component ratio and the standard thearubigins component ratio, and the theabrownins deviation ratio is the absolute value of the difference between the theabrownins component ratio and the standard theabrownins component ratio.

[0045] Specifically, in step S4, the process of adjusting the baking process parameters includes... Step S401: Calculate several ratio differences between the deviation ratio of each tea pigment and the corresponding preset deviation ratio of tea pigment. Step S402: Adjust the baking parameters based on the comparison results of each ratio difference and the preset difference.

[0046] It is understandable that the proportion of theaflavins and thearubigins in the second-grade dark roasted tea is higher than that in the third-grade dark roasted tea, while the proportion of theabrownins in the second-grade dark roasted tea is lower than that in the third-grade dark roasted tea.

[0047] It is understandable that the theaflavins ratio difference is the difference between the theaflavins deviation ratio and the preset theaflavins deviation ratio; the thearubigins ratio difference is the difference between the thearubigins deviation ratio and the preset thearubigins deviation ratio; and the theabrownins ratio difference is the difference between the theabrownins deviation ratio and the preset theabrownins deviation ratio.

[0048] It is understandable that since theaflavins and thearubigins are pre-roasting products of theabrownins in heavily roasted tea, the ratio difference between theaflavins and thearubigins is calculated by taking into account both and recorded as the tea precursor ratio difference. The tea precursor ratio difference is the average of the theaflavin ratio difference and the thearubigin ratio difference.

[0049] It is understandable that in heavily roasted tea leaves, the content of theabrownin is relatively high, while the content of theaflavins and thearubigins is relatively low. Therefore, the preferred preset difference value range for theabrownin is 3% to 5%, and the preset difference value range for tea precursors is 1% to 3%.

[0050] Specifically, if the difference in the proportion of theabrownins is greater than or equal to the preset difference in theabrownins and the difference in the proportion of tea precursors is greater than or equal to the preset difference in the proportion of tea precursors, then it is determined that the baking time and baking temperature should be increased. If the difference in the proportion of theabrownins is less than the preset difference in theabrownins and the difference in the proportion of tea precursors is greater than or equal to the preset difference in the proportion of tea precursors, then it is determined that the roasting time should be increased to increase the oxygen content. If the difference in the proportion of theabrownins is greater than or equal to the preset difference in theabrownins and the difference in the proportion of tea precursors is less than the preset difference in the proportion of tea precursors, then it is determined that the roasting time should be increased. If the difference in the proportion of theabrownins is less than the preset difference in theabrownins and the difference in the proportion of tea precursors is less than the preset difference in the proportion of tea precursors, then it is determined that the oxygen content should be increased.

[0051] Please see Figure 4 As shown, this is a logic diagram for determining whether the adjusted baking process parameters meet the preset standard according to an embodiment of the present invention. In step S5, the determination of whether the adjusted baking process parameters meet the preset standard based on the optimization deviation ratio includes, If the optimization deviation ratio is less than the preset optimization deviation ratio, it is determined that the adjustment of the roasting process parameters meets the preset standard and the reason for the formation of the third roasting grade of heavy roasted tea is the control deviation of the roasting process parameters. If the optimization deviation ratio is greater than or equal to the preset optimization deviation ratio, it is determined that the adjustment of the roasting process parameters does not meet the preset standard and that the cause of the third roasting grade of heavy roasted tea is a problem with the tea raw materials.

[0052] It is understandable that the longer the roasting time, the better the roasting effect of the heavily roasted tea. Therefore, the preset optimization deviation ratio is negatively correlated with the roasting time. The preferred preset optimization deviation ratio is 3% to 5%.

[0053] In a specific embodiment, the preset optimization deviation ratio is set to 3%. If the optimization deviation ratio is 1%, which is less than the preset optimization deviation ratio, it is determined that the adjustment of the roasting process parameters meets the preset standard and that the reason for the formation of the heavy roasted tea of ​​the third roasting grade is the control deviation of the roasting process parameters. If the optimization deviation ratio is 12%, which is greater than the preset optimization deviation ratio, it is determined that the adjusted roasting process parameters do not meet the preset standards and that the cause of the third roasting grade of heavy roasted tea is a problem with the tea raw materials.

[0054] Specifically, in step S6, the process of determining the tea pigment composition of the third-grade dark roasted tea based on the obtained tea liquor image of the third-grade dark roasted tea and the standard tea liquor image includes, Step S601: Obtain tea infusion parameters based on the tea infusion image of the third roasting grade of the heavily roasted tea leaves. Step S602: Obtain standard tea infusion parameters based on the standard tea infusion image; Step S603: Calculate the similarity value between the tea infusion parameters and the standard tea infusion parameters, and determine the tea pigment composition of the third roasting grade of dark roasted tea based on the similarity value.

[0055] The parameters of tea infusion include the mean value of tea infusion brightness, the standard deviation of tea infusion saturation, and the median value of the hue peak interval; the parameters of standard tea infusion include the standard tea infusion brightness value, the standard deviation of standard tea infusion saturation, and the median value of the standard hue peak interval.

[0056] In practice, third-grade and first-grade heavy-roasted tea leaves of the same quality were selected and brewed at the same time and temperature to prepare corresponding tea infusions. A transparent glass beaker containing the tea infusion was placed in front of a white soft-light background. Optionally, the background brightness was set to 500 lux to avoid strong light reflection or shadow interference. A high-definition industrial camera was used to vertically capture images of the tea infusion. Optionally, the distance between the camera lens and the surface of the tea infusion was fixed at 30cm, the aperture was set to F8, the shutter speed was 1 / 125s, and the ISO value was 100 to ensure that the shooting parameters were consistent each time. When shooting, the entire surface of the tea infusion and the lower half of the side wall of the beaker should be covered to avoid parameter deviations caused by shooting only a part. Three parallel images were obtained to reduce random errors.

[0057] The similarity value is determined based on the comparison results between the obtained tea infusion parameters and the standard tea infusion parameters. For example, the similarity value MP between the tea infusion parameter list YB=(YB1, YB2, ..., YBj, ..., YBm) and the standard tea infusion parameter list EB=(EB1, EB2, ..., EBj, ..., EBm); where j=1, 2, ..., m. Similarity value: MP=∑mj=1YBj*EBj / (sqrt(∑mj=1(YBj)2)*sqrt(∑mj=1(EBj)2)).

[0058] It is understandable that the parameters of tea soup include the brightness value, the standard deviation of tea soup saturation, and the median of the hue peak interval; the parameters of standard tea soup include the brightness value, the standard deviation of standard tea soup saturation, and the median of the standard hue peak interval, so m is taken as 3.

[0059] Understandably, the pigment composition of heavily roasted tea leaves of the third roasting grade is determined based on similarity values. The pigment composition of the heavily roasted tea leaves corresponding to the standard tea infusion parameter with the highest similarity value (greater than 0.9) is selected as the pigment composition of the third roasting grade tea leaves corresponding to that tea infusion parameter. If no heavily roasted tea leaves correspond to the standard tea infusion parameter with the highest similarity value (greater than 0.9), then the pigment composition of the third roasting grade tea leaves corresponding to that tea infusion parameter is quantitatively analyzed. The analysis results are stored in correspondence with the tea infusion parameter for use as a standard tea infusion parameter reference in the next analysis of pigment components.

[0060] Specifically, this invention compares the differences in tea pigments between first-grade and second-grade heavy-roasted tea leaves. It adjusts the roasting parameters of insufficiently roasted second-grade heavy-roasted tea leaves to obtain first-grade heavy-roasted tea leaves that meet the standards. This also allows for accurate quality control of the raw materials used for subsequent tea pigment extraction. Furthermore, based on the analysis of the adjusted second-grade heavy-roasted tea leaves, it determines whether sufficient roasting has been achieved, thus identifying the reasons for the formation of third-grade heavy-roasted tea leaves. If the analysis shows that the adjusted second-grade heavy-roasted tea leaves have achieved sufficient roasting, it indicates that adjusting the roasting process parameters can control the roasting process. The deviation in roasting, specifically the formation of the third-grade heavy roasted tea, is due to deviations in the control of roasting process parameters. If the adjustment results of the second-grade heavy roasted tea, after adjustment, do not achieve sufficient roasting, it indicates that the initial substances for forming tea pigments in the raw materials of the heavy roasted tea are insufficient. In other words, the formation of the third-grade heavy roasted tea is due to a problem with the tea raw materials. Therefore, for the third-grade tea whose formation is determined to be due to a problem with the tea raw materials, its tea soup image is extracted and compared with the standard tea soup image of the same tea. Based on the similarity value, the tea pigment composition of the third-grade heavy roasted tea is determined, further improving the accuracy of visual inspection of tea.

[0061] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for adjusting roasting process parameters based on low-temperature synergistic extraction of tea pigments, characterized in that, include: Based on the acquired dry tea image, it is determined whether the tea raw material is a first light roast tea. Based on the acquired tea soup image of non-first light roast tea, the non-first light roast tea is classified. Collect images of the leaf base of heavily roasted tea leaves, and determine the roasting grade of the heavily roasted tea leaves based on the leaf base images; Standard tea pigments were extracted from the first-grade dark roasted tea leaves to determine the proportion of standard tea pigment components and standard tea infusion images. The proportion of tea pigment components in the second-grade dark roasted tea leaves was then used to determine the tea pigment deviation proportion. Adjust the baking process parameters based on the ratio difference between the tea pigment deviation ratio and the corresponding preset tea pigment deviation ratio; The proportion of optimized tea pigment components in the heavily roasted tea produced after adjusting the roasting process parameters is extracted and combined with the proportion of standard tea pigment components to determine the optimization deviation ratio. Based on the optimization deviation ratio, it is determined whether the adjusted roasting process parameters meet the preset standard and the cause of the formation of the third roasting grade of heavily roasted tea is determined. The tea pigment composition of the third-grade dark roasted tea was determined based on the tea liquor image obtained from the third-grade dark roasted tea and the standard tea liquor image. The tea pigments include theaflavins, thearubigins, and theabrownins, and the tea pigment deviation ratios include the theaflavins deviation ratio, thearubigins deviation ratio, and theabrownins deviation ratio. The process of determining the deviation ratio of tea pigments includes: Under the condition that the roasting grade is determined to be the second roasting grade, the tea pigments in the corresponding heavily roasted tea leaves are extracted by ultrasonic low temperature method. The proportions of tea pigment components in the corresponding heavily roasted tea leaves were obtained by analyzing the tea pigments in the tea pigments using spectrophotometry. The deviation ratio of each tea pigment is determined by combining the proportion of the tea pigment components and the proportion of the standard tea pigment components. The process of determining the proportions of standard components of tea pigments includes: Under the condition that the roasting grade is determined to be the first roasting grade, the tea pigment in the corresponding heavily roasted tea leaves is extracted by ultrasonic low temperature method and recorded as the standard tea pigment; The proportions of standard tea pigment components were obtained by analyzing the standard tea pigments using spectrophotometry. The baking process parameters include baking temperature, baking time, and oxygen content.

2. The method for adjusting roasting process parameters based on low-temperature synergistic extraction of tea pigments according to claim 1, characterized in that, The process of determining whether the tea raw material is a first-light-roasted tea based on the obtained dry tea image includes the following steps: The tea leaves are laid flat on a white background to obtain an initial image of the dry tea. The initial dry tea image is cropped to obtain the dry tea image, and the dry tea image is denoised and converted to the HSV color model. The proportion of green pixel areas in the dry tea image is calculated and recorded as the light roasting ratio. Based on the light roasting ratio, it is determined whether the tea raw material is a first light roasted tea.

3. The method for adjusting roasting process parameters based on low-temperature synergistic extraction of tea pigments according to claim 2, characterized in that, The process of classifying the non-first light roasted tea leaves includes: Acquire images of tea infusions from non-first-light-roasted tea leaves and convert them to an HSV color model; The brightness value is calculated based on the tea soup image; The brightness value is compared with the first preset brightness value and the second preset brightness value respectively, and the non-first light roast tea is classified according to the comparison result; The classification of non-first light roast tea includes second light roast tea, medium roast tea, and heavy roast tea.

4. The method for adjusting roasting process parameters based on low-temperature synergistic extraction of tea pigments according to claim 3, characterized in that, The process of determining the baking grade based on the leaf base image includes: Remove the fully expanded tea leaves after brewing and lay them flat on a white background to obtain the initial leaf image; The initial leaf base image is cropped to obtain the leaf base image, and the leaf base image is converted to the HSV color model; The proportion of green pixel areas in the leaf base image is calculated and recorded as the leaf base ratio. The baking grade is determined based on the leaf base ratio. The leaf base image is an image that only retains the base of the tea leaves.

5. The method for adjusting roasting process parameters based on low-temperature synergistic extraction of tea pigments according to claim 4, characterized in that, The baking grade is determined based on the leaf-to-root ratio, wherein... If the leaf-to-root ratio is less than the first preset leaf-to-root ratio, then the baking grade is determined to be the first baking grade; If the leaf-to-root ratio is greater than or equal to the first preset leaf-to-root ratio and less than the second preset leaf-to-root ratio, then the baking grade is determined to be the second baking grade. If the leaf-to-root ratio is greater than or equal to the second preset leaf-to-root ratio, then the baking grade is determined to be the third baking grade; The first preset leaf-to-bottom ratio and the second preset leaf-to-bottom ratio are negatively correlated with the baking time, and the first preset leaf-to-bottom ratio is less than the second preset leaf-to-bottom ratio.

6. The method for adjusting roasting process parameters based on low-temperature synergistic extraction of tea pigments according to claim 5, characterized in that, The process of adjusting the baking process parameters includes: Calculate several ratio differences between the deviation ratio of each tea pigment and the corresponding preset deviation ratio of tea pigment; The baking process parameters are adjusted based on the comparison results between the stated ratio differences and the preset differences.

7. The method for adjusting roasting process parameters based on low-temperature synergistic extraction of tea pigments according to claim 6, characterized in that, Based on the optimized deviation ratio, it is determined whether the adjusted baking process parameters meet the preset standard. include, If the optimization deviation ratio is less than the preset optimization deviation ratio, it is determined that the adjustment of the roasting process parameters meets the preset standard and that the reason for the formation of the third roasting grade of heavy roasted tea is the control deviation of the roasting process parameters. If the optimization deviation ratio is greater than or equal to the preset optimization deviation ratio, it is determined that the adjusted roasting process parameters do not meet the preset standard and that the cause of the third roasting grade of heavy roasted tea is a problem with the tea raw materials.

8. The method for adjusting roasting process parameters based on low-temperature synergistic extraction of tea pigments according to claim 7, characterized in that, Assuming the cause of the third-grade heavy roasted tea is determined to be a problem with the tea raw materials, the process of determining the tea pigment composition of the third-grade heavy roasted tea based on the obtained tea infusion image of the third-grade heavy roasted tea and the standard tea infusion image includes the following steps: Tea infusion parameters are obtained based on the tea infusion image of the third roasting grade of dark roasted tea leaves. Obtain standard tea infusion parameters based on the standard tea infusion image; Calculate the similarity value between the tea infusion parameters and the standard tea infusion parameters, and determine the tea pigment composition of the third roasting grade of dark roasted tea based on the similarity value.

Images