A purple clay teapot seal image generation system

Abstract

The present invention relates to the technical field of image processing, and particularly to a purple clay teapot seal image generation system. The system includes an interaction module for obtaining a grayscale image and an HSV image of a purple clay teapot seal image; a feature extraction module for extracting features of the obtained purple clay teapot image and judging the processing intensity of the purple clay teapot image; a text enhancement module for analyzing a text enhancement scheme for the purple clay teapot seal image; a texture enhancement module for analyzing a texture enhancement scheme for the purple clay teapot seal image; a color adjustment module for analyzing a color adjustment scheme for the purple clay teapot seal image; an image generation module for outputting an image generation scheme to the user; a quality evaluation module for evaluating the quality of the generated image; and a quality monitoring module for monitoring the quality of the generated image within a monitoring period according to the results of the quality evaluation of the generated image within the monitoring period. The present invention improves the generation efficiency of the seal image.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to a purple clay teapot seal image generation system. Background Art

[0002] The purple clay teapot is one of the important art crafts in Chinese traditional culture, famous for its unique material and exquisite craftsmanship. The purple clay teapot seal is not only a symbol of identity and quality but also an important mark for art appreciation. The existing seal image generation technologies mainly rely on traditional image processing methods such as Fourier transform and Laplace transform, but there are deficiencies in dealing with complex textures and detailed information.

[0003] Chinese Patent Publication No. CN116311272A discloses a method for generating a seal image and a method for identifying a seal. It relates to the field of artificial intelligence. The method includes: obtaining an initial template image containing an elliptical seal contour, an organization name, and a bottom plate image, and determining the character size; for each character in the organization name, determining the target arc length corresponding to the character according to the preset total arc length and the target sequence position of the character; obtaining the mapping relationship between the arc length and the centrifugal angle of the elliptical seal contour, determining the target information of the character according to the target arc length and the mapping relationship, and arranging the character in the bottom plate image according to the target information to obtain a target bottom plate image matching the character; adding each character to the initial template image according to the target bottom plate image matched by each character to obtain a seal image matching the organization name. The present invention solves the technical problem of low annotation efficiency existing in using a real seal picture as a training sample for a seal recognition model in the existing technology during the seal recognition process; thus, it can be seen that when generating a seal image with this solution, the processing intensity of the seal image is not analyzed, resulting in a problem of low seal image generation efficiency. Summary of the Invention

[0004] The purpose of the present invention is to provide a purple clay teapot seal image generation system to solve at least one of the problems existing in the prior art.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions:

[0006] An interaction module for obtaining the grayscale image and HSV image of the purple clay teapot seal image;

[0007] A feature extraction module for extracting features from the grayscale image and HSV image of the obtained purple clay teapot seal image, and judging the processing intensity of the purple clay teapot seal image according to the extraction result;

[0008] A text enhancement module for analyzing the text enhancement scheme of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image;

[0009] A texture enhancement module for analyzing the texture enhancement scheme of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image;

[0010] A color adjustment module for analyzing the color adjustment scheme of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image;

[0011] An image generation module for outputting the text enhancement scheme, texture enhancement scheme and color adjustment scheme to the user as an image generation scheme;

[0012] A quality assessment module for constructing the detail similarity between the generated image and the purple clay teapot seal image, and evaluating the quality of the generated image according to the construction result of the detail similarity;

[0013] A quality monitoring module for monitoring the quality of the generated images within the monitoring period according to the results of the quality assessment of the generated images within the monitoring period.

[0014] Furthermore, the feature extraction module is provided with a sharpness analysis unit, a contrast analysis unit, a noise analysis unit, a richness analysis unit, a saturation analysis unit and a processing intensity judgment unit; the sharpness analysis unit constructs the image sharpness index n1 according to the grayscale image of the obtained purple clay teapot seal image, and sets: Among them, represents the result of the image I processed by the Laplacian operator, represents the variance of;

[0015] The contrast analysis unit constructs the image contrast index n2 according to the grayscale image of the obtained purple clay teapot seal image, and sets:

[0016]

[0017] Among them, H 均 is the average value of the pixels of each pixel point of the grayscale image of the purple clay teapot seal image, H(i) is the number of pixel points with the gray value of i, i represents the gray value, and N is the number of pixel points in the grayscale image;

[0018] The noise analysis unit divides the obtained grayscale image of the purple clay teapot seal image into E parts, and constructs the image noise index n3, and sets:

[0019]

[0020] Among them, Mj is the number of pixel points of the j-th part of the purple clay teapot seal image, Lm is the gray value of the m-th pixel point of the j-th part of the purple clay teapot seal image, and μj is the average value of the gray values of each pixel point of the j-th part of the purple clay teapot seal image.

[0021] Further, the richness analysis unit is used to construct the image richness index n4 based on the grayscale image of the purple clay teapot seal image, and it is set that:

[0022]

[0023] where Wq is the high-frequency coefficient of the q-th wavelet transform, and Q is the number of high-frequency coefficients;

[0024] The saturation analysis unit constructs the purple clay teapot seal image saturation index n5 based on the HSV image of the purple clay teapot seal image, and it is set that:

[0025]

[0026] where N is the number of pixel points in the HSV image of the purple clay teapot seal image, and St is the saturation value of the t-th pixel point;

[0027] The processing intensity judgment unit constructs the image processing intensity index P based on the image sharpness index, image contrast index, image noise index, image richness index and image saturation index, and it is set that: P = η × {x1 × min{max[(k1 - n1) / k1, 0], 1} + x2 × min{max[(k2 - n2) / k2, 0], 1} + x3 × min{max[(k3 - n3) / k3, 0], 1} + x4 × min{max[(k4 - n4) / k4, 0], 1} + x5 × min{max[(k5 - n5) / k5, 0], 1}}, where x1 is the sharpness weight, k1 is the preset sharpness threshold, x2 is the contrast weight, k2 is the preset contrast threshold, x3 is the noise weight, k3 is the preset noise threshold, x4 is the richness weight, k4 is the preset richness threshold, x5 is the saturation weight, k5 is the preset saturation threshold, and η is the preset proportionality coefficient.

[0028] Further, the text enhancement module analyzes the text sharpening intensity of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image, and sets the text sharpening intensity of the purple clay teapot seal image as R1, and it is set that R1 = P;

[0029] The text enhancement module analyzes the text contrast of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image, and sets the text contrast of the purple clay teapot seal image as R2, and it is set that R2 = 1 + exp(lgP),

[0030] The text enhancement module takes the analysis result of the text sharpening intensity of the purple clay teapot seal image and the analysis result of the text contrast as the text enhancement scheme.

[0031] Further, the texture enhancement module analyzes the texture detail enhancement intensity of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image, and sets the texture detail enhancement intensity of the purple clay teapot seal image as Z1, and sets Z1 = 3×P 2 - 2×P 3 ;

[0032] The texture enhancement module analyzes the texture contrast of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image, and sets the texture contrast of the purple clay teapot seal image as Z2, and sets Z2 = 0.8 + α×P, where α is the first preset adjustment coefficient;

[0033] The texture enhancement module takes the analysis result of the texture detail enhancement intensity and the analysis result of the texture contrast of the purple clay teapot seal image as the texture enhancement scheme.

[0034] Further, the color adjustment module analyzes the saturation of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image, and sets the saturation of the purple clay teapot seal image as D1, and sets D1 = 0.8 + β×P, where β is the second preset adjustment coefficient;

[0035] The color adjustment module analyzes the lightness of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image, and sets the lightness of the purple clay teapot seal image as D2, and sets D2 = 0.9 + γ×P, where β is the third preset adjustment coefficient;

[0036] The color adjustment module takes the analysis result of the saturation and the analysis result of the lightness of the purple clay teapot seal image as the color adjustment scheme.

[0037] Further, the quality assessment module is provided with a similarity analysis unit and a quality assessment unit. The similarity analysis unit is used to obtain the grayscale image of the generated image and construct the detail similarity G between the generated image and the purple clay teapot seal image, and set:

[0038] G=(2H 均 ×μA + C1)×(2σIA + C2) / [(H 均 2 + μA 2 + C1)×(σI 2 + σ

[0039] A 2 + C2)];

[0040] Among them, μA is the average value of the grayscale values of each pixel point in the generated image, σI 2 is the variance of the grayscale values of each pixel point in the purple clay teapot seal image, σA 2To generate the variance of the grayscale values of each pixel in the image, σIA is the covariance between the purple clay teapot seal image and the generated image, C1 is the first preset constant, and C2 is the second preset constant.

[0041] Further, the quality evaluation unit is used to compare the detail similarity G with a preset similarity to evaluate the quality of the generated image, where:

[0042] If G≥G1, the quality evaluation unit determines that the quality of the generated image is normal; if G<G1, the quality evaluation unit determines that the quality of the generated image is abnormal, where G1 is the preset similarity.

[0043] Further, the quality monitoring module is provided with a quality monitoring unit and a volatility analysis unit. The quality monitoring unit monitors the quality of the generated images within the monitoring period according to the results of the quality evaluation of the generated images within the monitoring period, and optimizes the analysis process of the image generation scheme for the next monitoring period according to the quality monitoring results, where:

[0044] If B0 / B1≤b, the quality monitoring unit determines that the quality of the generated images in the current monitoring period is qualified and does not perform optimization; if B0 / B1>b, the quality monitoring unit determines that the quality of the generated images in the current monitoring period is unqualified, and optimizes the analysis process of the image generation scheme for the next monitoring period, and sets the optimized preset proportional coefficient as η’. B0 is the number of generated images with abnormal quality within the monitoring period, B1 is the number of generated images within the monitoring period, and b is the preset abnormal ratio.

[0045] Further, the volatility analysis unit constructs a volatility coefficient BD based on the analysis results of the detail similarity of the generated images within the monitoring period, and sets and adjusts the optimization of the analysis process of the image generation scheme for the next monitoring period according to the analysis results, where:

[0046] If BD≤bd, the volatility analysis unit determines that the volatility of the detail similarity of the generated images in the current monitoring period is normal and does not perform adjustment; if BD>bd, the volatility analysis unit determines that the volatility of the detail similarity of the generated images in the current monitoring period is abnormal, and adjusts the optimization of the analysis process of the image generation scheme for the next monitoring period, and sets the adjusted preset abnormal ratio as b’. Gg is the detail similarity of the g-th generated image within the monitoring period.

[0047] The beneficial effects of the present invention are as follows: The feature extraction module quantifies multiple factors such as clarity, contrast, noise, richness, and saturation to better understand the overall quality of the image, and then makes precise adjustments. This comprehensive analysis ability not only improves the scientificity and rationality of image processing. The text enhancement module can ensure that the text on the purple clay teapot seal is clear and easy to read by enhancing the text, improving the efficiency and accuracy of information transmission. The text sharpening preserves fine text features, and the enhanced contrast makes the level and position of the text more prominent in the image. The texture enhancement module enables the delicate sense of the pattern to be reflected and can also provide users with a deeper visual experience. The color adjustment module can make the entire image visually more harmonious and beautiful by optimizing saturation and brightness, improving the user's image experience. The quality assessment module optimizes the image processing process by analyzing the similarity between the generated image and the original image to provide feedback, continuously improving the standard and quality of image generation, thereby improving the efficiency of seal image generation and enhancing the user experience. The quality monitoring module can adjust the processing parameters in real time by analyzing abnormal images, improving the success rate and quality level of image generation, ultimately improving the efficiency of seal image generation and enhancing the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0049] Figure 1 It is a schematic structural diagram of the purple clay teapot seal image generation system in this embodiment.

[0050] Figure 2 It is a schematic structural diagram of the feature extraction module in this embodiment.

[0051] Figure 3 It is a schematic structural diagram of the quality assessment module in this embodiment.

[0052] Figure 4 It is a schematic structural diagram of the quality monitoring module in this embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] In order to more clearly illustrate the present invention, the following further describes the present invention in conjunction with the preferred embodiments and the accompanying drawings. Similar components in the drawings are denoted by the same reference numerals. Those skilled in the art should understand that the content specifically described below is illustrative rather than restrictive, and should not be used to limit the protection scope of the present invention.

[0054] It should be noted that although terms such as first, second, and third may be used in the embodiments of the present application for description, these descriptions should not be limited to these terms. These terms are only used to distinguish the descriptions. For example, without departing from the scope of the embodiments of the present application, the first may also be referred to as the second, and similarly, the second may also be referred to as the first.

[0055] Please refer to Figure 1 as shown, which is a schematic structural diagram of the purple clay teapot seal image generation system in this embodiment. The system includes

[0056] an interaction module for obtaining the grayscale image and HSV image of the purple clay teapot seal image; in this embodiment, the acquisition methods of the grayscale image and HSV image of the purple clay teapot seal image are not specifically limited, and those skilled in the art can freely set them as long as the acquisition requirements of the grayscale image and HSV image of the purple clay teapot seal image are met. Among them, the grayscale image and HSV image of the purple clay teapot seal image can be obtained through OpenCV and PIL in Python, and the purple clay teapot seal image can be obtained through user interaction;

[0057] a feature extraction module for extracting features from the obtained grayscale image and HSV image of the purple clay teapot seal image, and judging the processing intensity of the purple clay teapot image according to the extraction results. The feature extraction module is connected to the interaction module; please refer to Figure 2 as shown, the feature extraction module includes

[0058] a clarity analysis unit for constructing an image clarity index n1 based on the obtained grayscale image of the purple clay teapot seal image to improve the accuracy of feature extraction of the purple clay teapot seal image,

[0059] The construction of the image clarity index n1 is as follows:

[0060] Among them, represents the result after the image I is processed by the Laplacian operator, represents the variance of; the clarity analysis unit enhances the adaptability to images of different qualities by constructing the image clarity index, and improves the robustness and stability of the system in practical applications.

[0061] Specifically, in this embodiment, the acquisition method of is not specifically limited, and those skilled in the art can freely set it as long as the acquisition requirements of are met. Among them, can be obtained using the convolution functions of OpenCV and scikit-image.

[0062] Please continue to refer to Figure 2. The feature extraction module further includes a contrast analysis unit, which constructs an image contrast index n2 based on the grayscale image of the purple clay teapot seal image to improve the accuracy of feature extraction of the purple clay teapot seal image. The contrast analysis unit is connected to the clarity analysis unit;

[0063] The construction of the image contrast index n2 is as follows:

[0064]

[0065] where H 均 is the average value of the pixels of each pixel point of the grayscale image of the purple clay teapot seal image, H(i) is the number of pixel points with a grayscale value of i, i represents the grayscale value, and N is the number of pixel points in the grayscale image; by analyzing the contrast of the image, the contrast analysis unit can better identify the details and features in the image. Contrast analysis can reveal the subtle structure and texture changes of the seal, thereby improving the accuracy of feature extraction.

[0066] Specifically, in this embodiment, the method for obtaining the number of pixel points with a grayscale value of i and the number of pixel points in the grayscale image is not specifically limited. Those skilled in the art can freely set it as long as the requirements for obtaining the number of pixel points with a grayscale value of i and the number of pixel points in the grayscale image are met. Among them, the number of pixel points with a grayscale value of i can be obtained by using the np.histogram() function in the NumPy library, and the number of pixel points in the grayscale image can be obtained by OpenCV.

[0067] Please continue to refer to Figure 2 Figure 2. The feature extraction module further includes a noise analysis unit, which divides the grayscale image of the obtained purple clay teapot seal image into E parts and constructs an image noise index n3 to improve the accuracy of feature extraction of the purple clay teapot seal image. The noise analysis unit is connected to the contrast analysis unit;

[0068] The construction of the image noise index n3 is as follows:

[0069]

[0070] where Mj is the number of pixel points of the j-th part of the purple clay teapot seal image, Lm is the grayscale value of the m-th pixel point of the j-th part of the purple clay teapot seal image, and μj is the average value of the grayscale values of each pixel point of the j-th part of the purple clay teapot seal image; by dividing the image grayscale image into E parts and calculating the image noise index n3, the noise analysis unit can more accurately identify and extract the features in the purple clay teapot seal image, which helps to reduce the interference of noise on feature extraction, thereby improving the accuracy of the extraction result.

[0071] Specifically, in this embodiment, there is no specific limitation on the acquisition methods of Lm and Mj. Those skilled in the art can freely set them as long as the acquisition requirements of Lm and Mj are met. Among them, Mj can be obtained through the size attribute of numpy, and Lm can be obtained through traversing the numpy array; in this embodiment, there is no specific limitation on the segmentation method of the grayscale image of the purple clay teapot seal image. Those skilled in the art can freely set it as long as the segmentation requirements of the grayscale image of the purple clay teapot seal image are met. Among them, the grayscale image of the purple clay teapot seal image can be equally divided into 8 parts.

[0072] Please continue to refer to Figure 2 As shown, the feature extraction module further includes a richness analysis unit, which is used to construct an image richness index n4 based on the obtained grayscale image of the purple clay teapot seal image to improve the accuracy of feature extraction of the purple clay teapot seal image. The richness analysis unit is connected to the noise analysis unit;

[0073] The construction of the image richness index n4 is as follows:

[0074]

[0075] Among them, Wq is the high-frequency coefficient of the qth wavelet transform, and Q is the number of high-frequency coefficients; through precise richness analysis, the richness analysis unit can reduce the interference of irrelevant information on feature extraction, thereby improving the overall efficiency of the system.

[0076] Specifically, the high-frequency coefficients of the wavelet transform in this embodiment are a set of coefficients obtained during the wavelet transform process, including the horizontal detail coefficient, vertical detail coefficient, and diagonal detail coefficient of the grayscale image. In this embodiment, there is no specific limitation on the acquisition methods of Wq and Q. Those skilled in the art can freely set them as long as the acquisition requirements of Wq and Q are met. Among them, Wq and Q can be obtained through the PyWavelets library in Python.

[0077] Please continue to refer to Figure 2 As shown, the feature extraction module further includes a saturation analysis unit, which is used to construct a purple clay teapot seal image saturation index n5 based on the obtained HSV image of the purple clay teapot seal image to improve the accuracy of feature extraction of the purple clay teapot seal image. The saturation analysis unit is connected to the richness analysis unit;

[0078] The construction of the image saturation index n5 is as follows:

[0079] Where N is the number of pixel points in the HSV image of the purple clay teapot seal image, and St is the saturation value of the t-th pixel point; the saturation analysis unit constructs a saturation index to more accurately capture the color information of the purple clay teapot seal image, especially those subtle color difference changes, which helps to more accurately extract and identify the features in the image, thereby improving the accuracy of image recognition.

[0080] Specifically, in this embodiment, the method for obtaining St is not specifically limited, and those skilled in the art can freely set it as long as the requirements for obtaining St are met. Among them, St can obtain the saturation value of the pixel point through the OpenCV library in the Python language. In OpenCV, the function cv2.cvtColor() is used to convert the image from the BGR color space to the HSV color space, and then the saturation value of each pixel point in the HSV image can be directly accessed.

[0081] Please continue to refer to Figure 2 As shown, the feature extraction module further includes a processing intensity judgment unit, which constructs an image processing intensity index P based on the image clarity index, image contrast index, image noise index, image richness index, and image saturation index to better understand the overall quality of the image, and then make precise adjustments. The processing intensity judgment unit is connected to the saturation analysis unit;

[0082] The construction of the processing intensity index P is as follows: P = η × {x1 × min{max[(k1 - n1) / k1, 0], 1} + x2 × min{max[(k2 - n2) / k2, 0], 1} + x3 × min{max[(k3 - n3) / k3, 0], 1} + x4 × min{max[(k4 - n4) / k4, 0], 1} + x5 × min{max[(k5 - n5) / k5, 0], 1}}, where x1 is the clarity weight, k1 is the preset clarity threshold, x2 is the contrast weight, k2 is the preset contrast threshold, x3 is the noise weight, k3 is the preset noise threshold, x4 is the richness weight, k4 is the preset richness threshold, x5 is the saturation weight, k5 is the preset saturation threshold, η is the preset proportionality coefficient, and x1 + x2 + x3 + x4 + x5 = 1; the processing intensity judgment unit quantifies multiple factors such as clarity, contrast, noise, richness, and saturation to better understand the overall quality of the image, and then make precise adjustments. This comprehensive analysis ability not only improves the scientificity and rationality of image processing, ultimately improves the efficiency of seal image generation, but also improves the user experience.

[0083] Specifically, in this embodiment, no specific limitations are imposed on the values of the clarity weight, the preset clarity threshold, the contrast weight, the preset contrast threshold, the noise weight, the preset noise threshold, the richness weight, the preset richness threshold, the saturation weight, the preset saturation threshold, and the preset proportionality coefficient. Those skilled in the art can freely set them as long as they meet the value requirements of the clarity weight, the preset clarity threshold, the contrast weight, the preset contrast threshold, the noise weight, the preset noise threshold, the richness weight, the preset richness threshold, the saturation weight, the preset saturation threshold, and the preset proportionality coefficient. Among them, the best value of k1 is 25, the best value of k2 is 60, the best value of k3 is 12, the best value of k4 is 0.6, the best value of k5 is 0.5, the best value of x1 is 0.25, the best value of x2 is 0.2, the best value of x3 is 0.18, the best value of x4 is 0.2, the best value of x5 is 0.17, and the best value of η is 0.95.

[0084] Please continue to refer to Figure 1 As shown, the system further includes a text enhancement module, which is connected to the feature extraction module. It is used to analyze the text sharpening intensity of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image, so as to retain fine text features, and set the text sharpening intensity of the purple clay teapot seal image as R1, and set R1 = P;

[0085] The text enhancement module analyzes the text contrast of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image, so as to make the level and position of the text in the image more prominent, and set the text contrast of the purple clay teapot seal image as R2, and set R2 = 1 + exp(lgP),

[0086] The text enhancement module takes the analysis results of the text sharpening intensity and the text contrast of the purple clay teapot seal image as the text enhancement scheme; through the enhancement of the text, the text on the purple clay teapot seal can be ensured to be clear and easy to read, improving the efficiency and accuracy of information transmission. The text sharpening retains fine text features, while the enhanced contrast makes the level and position of the text in the image more prominent, ultimately improving the efficiency of seal image generation and the user experience.

[0087] Please continue to refer to Figure 1 As shown, the system further includes a texture enhancement module, which is connected to the text enhancement module; it is used to analyze the texture detail enhancement intensity of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image, so as to reflect the fineness of the pattern, and set the texture detail enhancement intensity of the purple clay teapot seal image as Z1, and set Z1 = 3×P 2 -2×P 3 ;

[0088] The texture enhancement module analyzes the texture contrast of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image to provide users with a deeper visual experience, and sets the texture contrast of the purple clay teapot seal image as Z2, where Z2 = 0.8 + α × P, α is the first preset adjustment coefficient, and 0.3 < α < 0.4;

[0089] The texture enhancement module takes the analysis results of the texture detail enhancement intensity and the texture contrast of the purple clay teapot seal image as the texture enhancement scheme; the texture enhancement module can reflect the meticulousness of the pattern and also provide users with a deeper visual experience, ultimately improving the efficiency of seal image generation and the user experience.

[0090] Specifically, in this embodiment, the setting of the first preset adjustment coefficient is not specifically limited, and those skilled in the art can freely set it as long as the setting requirements of the first preset adjustment coefficient are met. Among them, the optimal value of α is 0.32.

[0091] Please continue to refer to Figure 1 As shown, the system further includes a color adjustment module, which is connected to the texture enhancement module; it is used to analyze the saturation of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image to make the entire image visually more harmonious and beautiful, and sets the saturation of the purple clay teapot seal image as D1, where D1 = 0.8 + β × P, β is the second preset adjustment coefficient, and 0.28 < β < 0.4;

[0092] The color adjustment module analyzes the lightness of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image to improve the user's image experience, and sets the lightness of the purple clay teapot seal image as D2, where D2 = 0.9 + γ × P, β is the third preset adjustment coefficient, and 0.15 < γ < 0.2;

[0093] The color adjustment module takes the analysis results of the saturation and lightness of the purple clay teapot seal image as the color adjustment scheme; the color adjustment module can make the entire image visually more harmonious and beautiful by optimizing the saturation and lightness, improve the user's image experience, ultimately improve the efficiency of seal image generation, and improve the user experience at the same time.

[0094] Specifically, in this embodiment, the settings of the second preset adjustment coefficient and the third preset adjustment coefficient are not specifically limited, and those skilled in the art can freely set them as long as the setting requirements of the second preset adjustment coefficient and the third preset adjustment coefficient are met. Among them, the optimal value of β is 0.35, and the optimal value of γ is 0.18.

[0095] Please continue to refer to Figure 1As shown, the system further includes an image generation module, which is used to output text enhancement schemes, texture enhancement schemes, and color adjustment schemes to the user as image generation schemes. The image generation module is connected to the color adjustment module.

[0096] Please continue to refer to Figure 1 As shown, the system further includes a quality assessment module, which is used to construct the detail similarity between the generated image and the purple clay teapot seal image, and evaluate the quality of the generated image according to the construction result of the detail similarity. The quality assessment module is connected to the image generation module; Please refer to Figure 3 As shown, the quality assessment module includes a similarity analysis unit, which obtains the grayscale image of the generated image and constructs the detail similarity G between the generated image and the purple clay teapot seal image to provide feedback for optimizing the image processing process;

[0097] The construction of the detail similarity G is as follows:

[0098] G = (2H 均 × μA + C1) × (2σIA + C2) / [(H 均 2 + μA 2 + C1) × (σI 2 + σ

[0099] A 2 + C2)];

[0100]

[0101] Where vy is the grayscale value of the y-th pixel in the purple clay teapot seal image, fy is the grayscale value of the y-th pixel in the generated image, μA is the average grayscale value of each pixel in the generated image, σI 2 is the variance of the grayscale values of each pixel in the purple clay teapot seal image, σA 2 is the variance of the grayscale values of each pixel in the generated image, σIA is the covariance between the purple clay teapot seal image and the generated image, C1 is the first preset constant, and C2 is the second preset constant; The similarity analysis unit optimizes the image processing process by analyzing the similarity between the generated image and the original image to continuously improve the standard and quality of image generation, thereby improving the efficiency of seal image generation and enhancing the user experience.

[0102] Specifically, the generated image in this embodiment is the purple clay teapot seal image processed by the image generation scheme; the generated image in this embodiment is obtained through user interaction, and the acquisition method of the relevant parameters of the generated image is the same as that of the purple clay teapot seal image parameters. This embodiment will not elaborate, and the specific values of C1 and C2 are not specifically limited in this embodiment. Those skilled in the art can freely set them as long as the value requirements of C1 and C2 are met. Among them, when the image is an 8-bit image, the best value of C1 is (0.01×255)², and the best value of C2 is (0.03×255)².

[0103] Please continue to refer to Figure 3 As shown, the quality assessment module further includes a quality assessment unit, which compares the detail similarity G with a preset similarity to assess the quality of the generated image. The quality assessment unit is connected to the similarity analysis unit;

[0104] The assessment process of the quality of the generated image is as follows:

[0105] When G≥G1, the quality assessment unit determines that the quality of the generated image is normal;

[0106] When G<G1, the quality assessment unit determines that the quality of the generated image is abnormal;

[0107] Among them, G1 is the preset similarity.

[0108] Specifically, the setting of the preset similarity is not specifically limited in this embodiment. Those skilled in the art can freely set it as long as the setting requirements of the preset similarity are met. Among them, the best value of G1 is 0.8.

[0109] Please continue to refer to Figure 1 As shown, the system further includes a quality monitoring module, which is used to monitor the quality of the generated image within the monitoring period according to the results of the quality assessment of the generated image within the monitoring period. The quality monitoring module is connected to the quality assessment module; Please refer to Figure 4 As shown, the quality monitoring module includes a quality monitoring unit, which is used to monitor the quality of the generated image within the monitoring period according to the results of the quality assessment of the generated image within the monitoring period, analyze abnormal images, and optimize the analysis process of the image generation scheme in the next monitoring period according to the quality monitoring results to adjust the processing parameters in real time;

[0110] The optimization scheme of the analysis process of the image generation scheme in the next monitoring period is as follows:

[0111] When B0 / B1≤b, the quality monitoring unit determines that the quality of the generated image in the current monitoring period is qualified and no optimization is performed;

[0112] When B0 / B1 > b, the quality monitoring unit determines that the quality of the images generated in the current monitoring period is unqualified, and optimizes the analysis process of the image generation scheme for the next monitoring period. The optimized preset proportionality coefficient is set as η’, and η’ = η × [1 - (e (B0 / B1-b) / (B0 / B1+b)-1 ) / (e - 1)], where e is the natural logarithm, B0 is the number of images with abnormal quality generated during the monitoring period, B1 is the number of images generated during the monitoring period, and b is the preset abnormal ratio. Through the analysis of abnormal images, the quality monitoring unit can adjust the processing parameters in real time, improve the success rate and quality level of image generation, ultimately improve the efficiency of seal image generation, and at the same time improve the user experience.

[0113] Specifically, in this embodiment, the setting of the monitoring period is not specifically limited, and those skilled in the art can freely set it as long as it meets the requirements for setting the monitoring period. For example, the monitoring period can be set to 20 days, 30 days, etc. In this embodiment, the setting of the preset abnormal ratio is not specifically limited, and those skilled in the art can freely set it as long as it meets the requirements for setting the preset abnormal ratio. The best value of b is 0.05.

[0114] Please continue to refer to Figure 4 As shown, the quality monitoring module further includes a volatility analysis unit, which performs volatility analysis on the analysis results of the detail similarity of the images generated during the monitoring period to identify potential quality problems, and adjusts the optimization of the analysis process of the image generation scheme for the next monitoring period according to the analysis results. The volatility analysis unit is connected to the quality monitoring unit.

[0115] The construction process of the volatility coefficient BD is as follows:

[0116]

[0117] The adjustment scheme for the optimization of the analysis process of the image generation scheme for the next monitoring period is as follows:

[0118] When BD ≤ bd, the volatility analysis unit determines that the volatility of the detail similarity of the images generated during the current monitoring period is normal and does not make any adjustments.

[0119] When BD > bd, the volatility analysis unit determines that the volatility of the detail similarity of the images generated during the current monitoring period is abnormal, and adjusts the optimization of the analysis process of the image generation scheme for the next monitoring period. The adjusted preset abnormal ratio is set as b’, and b’ = b × exp[-(BD - bd)], where bd is the preset volatility coefficient, Gg is the detail similarity of the g-th generated image during the monitoring period, and G 均 is the average value of the detail similarity of the images generated during the monitoring period. The volatility analysis unit can identify potential quality problems and optimize abnormal situations in the image generation process by monitoring the volatility of the detail similarity. Volatility monitoring can ensure the consistency and stability of image generation, ultimately improving the efficiency of seal image generation and enhancing the user experience.

[0120] Specifically, in this embodiment, no specific limitation is imposed on the setting of the preset volatility coefficient, and those skilled in the art can freely set it as long as the setting requirements of the preset volatility coefficient are met. Among them, the optimal value of bd is 0.15.

[0121] Specifically, the purple clay teapot seal image generation system described in this embodiment is applied to the field of image processing. By performing multi-dimensional feature analysis on the original image of the purple clay teapot seal, it realizes the intelligent judgment of the processing intensity. The system adaptively generates text enhancement, texture enhancement, and color adjustment schemes according to the processing intensity, and conducts real-time evaluation and monitoring of the generated image quality, improving the generation efficiency of the seal image.

[0122] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, rather than limitations on the implementation manners of the present invention. For those of ordinary skill in the art, based on the above description, other different forms of changes or modifications can be made. It is impossible to enumerate all the implementation manners here. Any obvious changes or modifications derived from the technical solutions of the present invention still fall within the protection scope of the present invention.

Claims

1. A purple clay teapot seal image generation system, characterized in that: include, Interaction module, used to obtain the grayscale image and HSV image of the purple clay teapot seal image; A feature extraction module is used to extract features from the grayscale image and the HSV image of the acquired purple clay teapot seal image, and judge the processing intensity of the purple clay teapot seal image according to the extraction result, including: constructing an image processing intensity index P according to an image clarity index n1, an image contrast index n2, an image noise index n3, an image richness index n4, and an image saturation index n5; The treatment intensity index P is constructed as follows: P = η×{x1×min{max[(k1-n1) / k1,0],1}+x2×min{max[(k2-n2) / k2,0],1}+x3×min{max[(k3-n3) / k3,0],1}+x4×min{max[(k4-n4) / k4,0],1}+x5×min{max[(k5-n5) / k5,0],1}}, wherein x1 is the clarity weight, k1 is the preset clarity threshold, x2 is the contrast weight, k2 is the preset contrast threshold, x3 is the noise weight, k3 is the preset noise threshold, x4 is the richness weight, k4 is the preset richness threshold, x5 is the saturation weight, k5 is the preset saturation threshold, and η is the preset proportionality coefficient; The text enhancement module is used to analyze the text enhancement scheme of the purple clay teapot seal image according to the processing strength of the purple clay teapot image, including: analyzing the text sharpening strength and text contrast of the purple clay teapot seal image according to the processing strength index P of the purple clay teapot image, and using the analysis results of the text sharpening strength and text contrast of the purple clay teapot seal image as the text enhancement scheme; The texture enhancement module is used to analyze the texture enhancement scheme of the purple clay teapot seal image according to the processing strength of the purple clay teapot image, including: analyzing the texture detail enhancement strength and texture contrast of the purple clay teapot seal image according to the processing strength index P of the purple clay teapot image, and using the analysis results of the texture detail enhancement strength and the texture contrast of the purple clay teapot seal image as the texture enhancement scheme; The color adjustment module is used to analyze the color adjustment scheme of the purple clay teapot seal image according to the processing strength of the purple clay teapot image, including: analyzing the saturation and brightness of the purple clay teapot seal image according to the processing strength index P of the purple clay teapot image, and using the analysis results of the saturation and brightness of the purple clay teapot seal image as the color adjustment scheme; An image generation module, used to output the text enhancement scheme, the texture enhancement scheme and the color adjustment scheme as an image generation scheme to the user; A quality assessment module is used to construct the detail similarity between the generated image and the purple clay teapot seal image, and to evaluate the quality of the generated image based on the result of the detail similarity construction; The quality monitoring module is used to monitor the quality of the images generated within the monitoring period according to the results of the image quality assessment generated within the monitoring period.

2. The purple clay teapot seal image generation system according to claim 1, characterized in that: The feature extraction module is provided with a clarity analysis unit, a contrast analysis unit, a noise analysis unit, a richness analysis unit, a saturation analysis unit and a processing intensity judgment unit; the clarity analysis unit constructs an image clarity index n1 according to the grayscale image of the purple clay teapot seal image obtained, and the construction of the image clarity index n1 is as follows: in, It represents the result after the image I is processed by the Laplace operator. express The variance of The contrast analysis unit constructs an image contrast index n2 according to the acquired grayscale image of the purple clay teapot seal image. The image contrast index n2 is constructed as follows: Among them, H 均 is the average value of each pixel in the grayscale image of the purple clay teapot seal image, H(i) is the number of pixels with grayscale value i, i represents the grayscale value, and N is the number of pixels in the grayscale image; The noise analysis unit divides the grayscale image of the obtained purple clay teapot seal image into E parts, and constructs an image noise index n3, and the image noise index n3 is constructed as follows: Among them, Mj is the number of pixels of the j-th purple clay teapot seal image, Lm is the grayscale value of the m-th pixel of the j-th purple clay teapot seal image, and μj is the average grayscale value of each pixel of the j-th purple clay teapot seal image.

3. The purple clay teapot seal image generation system according to claim 2, characterized in that: The richness analysis unit is used to construct an image richness index n4 according to the grayscale image of the acquired purple clay teapot seal image, and the image richness index n4 is constructed as follows: Where Wq is the high-frequency coefficient of the qth wavelet transform, and Q is the number of high-frequency coefficients; The saturation analysis unit constructs a saturation index n5 of the purple clay teapot seal image according to the acquired HSV image of the purple clay teapot seal image. The image saturation index n5 is constructed as follows: Among them, N1 is the number of pixels in the HSV image of the purple clay teapot seal image, and St is the saturation value of the tth pixel; The processing intensity judgment unit constructs an image processing intensity index P according to an image clarity index, an image contrast index, an image noise index, an image richness index and an image saturation index.

4. The purple clay teapot seal image generation system according to claim 3, characterized in that: The text enhancement module analyzes the text sharpening strength of the purple clay teapot seal image according to the processing strength of the purple clay teapot image, and sets the text sharpening strength of the purple clay teapot seal image to R1, setting R1=P; The text enhancement module analyzes the text contrast of the purple clay teapot seal image according to the processing strength of the purple clay teapot image, and sets the text contrast of the purple clay teapot seal image to R2, setting R2=1+exp(lgP). The text enhancement module uses the analysis result of the text sharpening intensity of the purple clay teapot seal image and the analysis result of the text contrast as a text enhancement scheme.

5. The purple clay teapot seal image generation system according to claim 4, characterized in that: The texture enhancement module analyzes the texture detail enhancement strength of the purple clay teapot seal image according to the processing strength of the purple clay teapot image, and sets the texture detail enhancement strength of the purple clay teapot seal image to Z1, setting Z1=3×P 2 -2×P 3 ; The texture enhancement module analyzes the texture contrast of the purple clay teapot seal image according to the processing strength of the purple clay teapot image, and sets the texture contrast of the purple clay teapot seal image to Z2, setting Z2=0.8+α×P, where α is a first preset adjustment coefficient; The texture enhancement module uses the analysis result of the texture detail enhancement strength and the analysis result of the texture contrast of the purple clay teapot seal image as a texture enhancement scheme.

6. The purple clay teapot seal image generation system according to claim 5, characterized in that: The color adjustment module analyzes the saturation of the purple clay teapot seal image according to the processing intensity of the purple clay teapot image, and sets the saturation of the purple clay teapot seal image to D1, setting D1=0.8+β×P, where β is a second preset adjustment coefficient; The color adjustment module analyzes the lightness of the purple clay teapot seal image according to the processing strength of the purple clay teapot image, and sets the lightness of the purple clay teapot seal image to D2, setting D2=0.9+γ×P, where γ is a third preset adjustment coefficient; The color adjustment module uses the analysis results of the saturation and brightness of the purple clay teapot seal image as a color adjustment solution.

7. The purple clay teapot seal image generation system according to claim 6, characterized in that: The quality assessment module is provided with a similarity analysis unit and a quality assessment unit. The similarity analysis unit is used to obtain a grayscale image of the generated image and construct a detail similarity G between the generated image and the purple clay pot seal image. The setting is: G=(2H 均 ×μA+C1)×(2σIA+C2) / [(H 均 2 +μA 2 +C1)×(σI 2 +s A 2 +C2)]; Among them, μA is the average gray value of each pixel in the generated image, σI 2 is the variance of the grayscale value of each pixel in the purple clay pot seal image, σA 2 is the variance of the grayscale value of each pixel in the generated image, σIA is the covariance between the purple clay teapot seal image and the generated image, C1 is the first preset constant, and C2 is the second preset constant.

8. The purple clay teapot seal image generation system according to claim 7, characterized in that: The quality assessment unit is used to compare the detail similarity G with a preset similarity to assess the quality of the generated image, wherein: If G≥G1, the quality assessment unit determines that the quality of the generated image is normal; if G<G1, the quality assessment unit determines that the quality of the generated image is abnormal, wherein G1 is a preset similarity.

9. The purple clay teapot seal image generation system according to claim 8, characterized in that: The quality monitoring module is provided with a quality monitoring unit and a volatility analysis unit. The quality monitoring unit monitors the quality of the image generated in the monitoring period according to the result of the image quality evaluation generated in the monitoring period, and optimizes the analysis process of the image generation scheme of the next monitoring period according to the quality monitoring result, wherein: If B0 / B1≤b, the quality monitoring unit determines that the quality of the image generated in the current monitoring cycle is qualified and no optimization is performed; if B0 / B1>b, the quality monitoring unit determines that the quality of the image generated in the current monitoring cycle is unqualified, and optimizes the analysis process of the image generation plan for the next monitoring cycle, and sets the optimized preset proportional coefficient to η', where B0 is the number of images with abnormal quality generated in the monitoring cycle, B1 is the number of images generated in the monitoring cycle, and b is the preset abnormality ratio.

10. The purple clay teapot seal image generation system according to claim 9, characterized in that: The fluctuation analysis unit constructs the fluctuation coefficient BD based on the analysis results of the image detail similarity generated during the monitoring period, and sets And according to the analysis results, the optimization of the analysis process of the image generation scheme for the next monitoring cycle is adjusted, including: If BD≤bd, the volatility analysis unit determines that the volatility of the detail similarity of the generated images in the current monitoring period is normal and no adjustment is made; if BD>bd, the volatility analysis unit determines that the volatility of the detail similarity of the generated images in the current monitoring period is abnormal, and adjusts the optimization of the image generation scheme analysis process for the next monitoring period, and sets the adjusted preset abnormal ratio to b', bd is the preset volatility coefficient, Gg is the detail similarity of the g-th generated image in the monitoring period, and G 均 Generate the mean value of image detail similarity during the monitoring period.

Images