Purple sand ware identification method and system based on image processing

Abstract

The invention relates to the technical field of image processing, in particular to a purple sand ware identification method and system based on image processing, and the system comprises a data collection module which is used for collecting purple sand ware images and environment data; the preprocessing module is used for performing color normalization processing on the purple sand device image to obtain a corrected image; the region division module is used for performing region division on the corrected image according to the gradient of each pixel point of the corrected image in the horizontal direction and the gradient of each pixel point in the vertical direction so as to obtain a purple sand device region; the color feature analysis module is used for constructing a color coefficient of the purple sand ware area; the textural feature analysis module is used for analyzing textural features of the purple sand ware area; the gloss characteristic analysis module is used for analyzing the gloss characteristics of the purple sand ware area; and the judgment module is used for analyzing the pug type of the purple sand ware. According to the method, the accuracy of purple sand ware pug type analysis is effectively improved, and the identification efficiency of the purple sand ware is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a purple clay wares identification method and system based on image processing. Background Art

[0002] Zisha ware is a treasure of traditional Chinese arts and crafts. Its unique clay composition and craftsmanship imbue each piece with unique artistic value. Zisha ware's clays primarily include purple, red, and blue clays, each with distinct variations in color, texture, and luster. Therefore, accurately identifying the clay type of a piece is crucial for assessing its craftsmanship, collectible value, and authenticity.

[0003] Chinese Patent Publication No. CN110208303A discloses a method for detecting purple sandware, comprising the following steps: testing a standard raw material using energy dispersive X-ray fluorescence spectroscopy to obtain the fluorescence intensity of each element to be tested in the standard raw material, calculating the total fluorescence intensity of all elements to be tested and the relative proportion of each fluorescence intensity in the total fluorescence intensity; testing a sample to be tested using energy dispersive X-ray fluorescence spectroscopy to obtain the fluorescence intensity of each element to be tested in the sample to be tested, calculating the total fluorescence intensity of all elements to be tested and the relative proportion of each fluorescence intensity in the total fluorescence intensity; comparing the relative proportion of each element to be tested in the standard raw material with the relative proportion of each element to be tested in the sample to be tested. The above-mentioned method for detecting purple sandware can detect whether chemical raw materials are added to the sample to be tested, and the sample to be tested does not need to be damaged during the testing process; it can be seen that this scheme does not analyze the type of purple sandware clay when identifying purple sandware, resulting in a low efficiency in identifying purple sandware. Summary of the Invention

[0004] The object of the present invention is to provide a purple clay ware identification system based on image processing to solve at least one of the problems existing in the prior art.

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

[0006] A purple sandware identification system based on image processing, comprising:

[0007] Data acquisition module, used to collect images of purple clay pots and environmental data;

[0008] A pre-processing module is used to perform color normalization on the purple sandware image to obtain a corrected image;

[0009] A region division module is used to divide the corrected image into regions according to the horizontal gradient and the vertical gradient of each pixel point of the corrected image to obtain the purple sandware region;

[0010] The color feature analysis module is used to extract the color features of the HSV image of the purple sandware and construct the color coefficient of the purple sandware area based on the extraction results;

[0011] A texture feature analysis module is used to analyze the texture state of the purple sand ware area according to the contrast of the purple sand ware area and the image grayscale entropy of the purple sand ware area, and analyze the texture features of the purple sand ware area according to the analysis results;

[0012] The gloss feature analysis module is used to classify pixel types and analyze the gloss features of the purple sandware area based on the classification results;

[0013] The judgment module is used to analyze the type of purple clay material based on the construction results of the color coefficient of the purple clay area, texture characteristics, gloss characteristics and the collected ambient light intensity and ambient humidity.

[0014] Furthermore, the region division module calculates the gradient amplitude Pi of the i-th pixel point of the corrected image according to the gradient Tix in the horizontal direction and the gradient Tiy in the vertical direction of the i-th pixel point of the corrected image, and analyzes the relative position of the pixel point according to the gradient amplitude Pi of the i-th pixel point of the corrected image and the preset gradient amplitude threshold p0, where the relative position of the pixel point includes non-edge pixels and edge pixels;

[0015] The region division module uses a curve formed by connecting the edge pixel points as the outline of the purple sand ware, and uses the area within the outline of the purple sand ware as the purple sand ware area.

[0016] Furthermore, the color feature analysis module sets the hue of the HSV image of the purple sandware to Hu, and sets the hue coefficient to SD;

[0017] The color feature analysis module sets the saturation of the HSV image of the purple sandware to BH and the saturation coefficient to BHD;

[0018] The color feature analysis module sets the color coefficient of the purple sandware as Y, setting Y=r1×SD+r2×BHD;

[0019] Among them, r1 is the hue weight and r2 is the saturation weight.

[0020] Furthermore, the texture feature analysis module includes a contrast analysis unit, which is used to analyze the contrast state of the purple sandware area according to the contrast d0 of the purple sandware area, and construct a contrast coefficient according to the analysis result, and the contrast coefficient includes D1, D2 and D3;

[0021] The texture feature analysis module also includes a grayscale entropy analysis unit, which is used to analyze the grayscale entropy state of the purple sand ware area based on the image grayscale entropy s0 and the preset grayscale entropy s1 of the purple sand ware area. The grayscale entropy state of the purple sand ware area includes a normal state and an abnormal state. If the grayscale entropy state of the purple sand ware area is an abnormal state, the grayscale entropy coefficient is set to S1. If the grayscale entropy state of the purple sand ware area is a normal state, the grayscale entropy coefficient is set to S2.

[0022] Furthermore, the texture feature analysis module also includes a texture analysis unit, which is used to analyze the texture state of the purple clay ware area according to the construction result of the contrast coefficient. The texture state of the purple clay ware area includes a normal state and an abnormal state. If the texture state of the purple clay ware area is an abnormal state, the texture feature of the purple clay ware area is set to WT1. If the texture state of the purple clay ware area is a normal state, the texture feature of the purple clay ware area is set to WT2.

[0023] Furthermore, the gloss feature analysis module includes a type analysis unit, which is used to classify pixel types according to the grayscale value Lk of the kth pixel in the purple sandware area and a preset grayscale value threshold YZ. The pixel types include type I pixels and type II pixels.

[0024] The gloss feature analysis module further includes a gloss feature analysis unit, which is used to calculate the average grayscale value PG of the two types of pixel points, and analyze the gloss coefficient GX of the purple sandware according to the average grayscale value PG of the two types of pixel points and the average value Pu of the pixel points in the purple sandware area;

[0025] The gloss feature analysis unit analyzes the gloss feature of the purple clay ware area according to the gloss coefficient GX of the purple clay ware and the preset gloss coefficient GX0. If GX<GX0, the gloss feature analysis unit sets the gloss feature of the purple clay ware area to GT1; otherwise, the gloss feature analysis unit sets the gloss feature of the purple clay ware area to GT2.

[0026] Furthermore, the judgment module includes a judgment unit, which is used to analyze the type of purple sandware clay material according to the construction result of the color coefficient of the purple sandware area, the texture characteristics of the purple sandware area, and the gloss characteristics of the purple sandware area, wherein:

[0027] If w1×the color coefficient of the purple sand ware region+w2×the texture feature of the purple sand ware region+w3×the gloss feature of the purple sand ware region≤u0, the judgment unit determines that the purple sand ware clay is green clay; otherwise, the judgment unit determines that the purple sand ware clay is not green clay;

[0028] Among them, w1 is the color weight, w2 is the texture weight, w3 is the gloss weight, and u0 is the preset judgment threshold.

[0029] Furthermore, the judgment module includes a light analysis unit, which is used to analyze the light state according to the collected ambient light intensity g0 and the preset light intensity g1, and the light state includes a normal state and a strong light state;

[0030] The illumination analysis unit processes the analysis process of the purple clay ware type when the illumination state is a strong light state, and sets the preset judgment threshold after processing as u1.

[0031] Furthermore, the judgment module further includes a humidity analysis unit, which is used to analyze the humidity state according to the collected ambient humidity hs0 and the preset humidity hs1, and the humidity state includes a normal state and a high humidity state;

[0032] The humidity analysis unit processes the analysis process of the illumination state when the humidity state is a high humidity state, and sets the preset illumination intensity after processing to g2.

[0033] On the other hand, the present invention also provides a method for identifying purple sandware based on image processing, comprising:

[0034] Step S1, collecting purple sandware images and environmental data;

[0035] Step S2, performing color normalization processing on the purple sandware image to obtain a corrected image;

[0036] Step S3, dividing the corrected image into regions according to the horizontal gradient and the vertical gradient of each pixel point of the corrected image to obtain purple sandware regions;

[0037] Step S4, extracting color features from the HSV image of the purple sandware, and constructing a color coefficient of the purple sandware area based on the extraction results;

[0038] Step S5, analyzing the texture state of the purple sand ware region according to the contrast of the purple sand ware region and the image grayscale entropy of the purple sand ware region, and analyzing the texture characteristics of the purple sand ware region according to the analysis results;

[0039] Step S6, classifying the pixel types and analyzing the gloss characteristics of the purple sandware area based on the classification results;

[0040] Step S7: analyzing the type of purple sand ware clay material based on the color coefficient construction results, texture characteristics, gloss characteristics, and collected ambient light intensity and ambient humidity of the purple sand ware area.

[0041] The beneficial effects of the present invention are as follows: the data acquisition module provides a reliable basis for subsequent analysis through high-resolution image acquisition and environmental data recording, ensuring the accuracy of the data. In the preprocessing module, the color normalization process eliminates the influence of different shooting conditions on the color, making the image analysis more consistent; the region division module accurately extracts the outline of the purple clay ware, so that the subsequent analysis is only focused on the core area of the purple clay ware; the color feature analysis module accurately captures the unique color of the purple clay ware by constructing color coefficients, while the texture feature analysis module reveals the details of the surface process of the purple clay ware through the calculation of contrast and grayscale entropy, enhancing the objectivity of the identification; the gloss feature analysis module deeply analyzes the surface characteristics of the purple clay ware through the division of pixel types and the calculation of gloss coefficients, providing more dimensional information for judging authenticity; the judgment module comprehensively analyzes the color, texture and gloss features, combined with the ambient light and humidity, to ensure that accurate judgments can still be made under different conditions. Through the collaborative work of the above modules, the entire system can achieve efficient and accurate purple clay ware clay material identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 Schematic diagram of the structure of the purple clayware identification system based on image processing in this embodiment.

[0044] Figure 2 Schematic diagram of the structure of the texture feature analysis module of this embodiment.

[0045] Figure 3 Schematic diagram of the structure of the gloss feature analysis module of this embodiment.

[0046] Figure 4 This is a schematic diagram of the structure of the judgment module in this embodiment.

[0047] Figure 5 Schematic diagram of the flow of the purple clay ware identification method based on image processing in this embodiment. DETAILED DESCRIPTION

[0048] In order to more clearly illustrate the present invention, the present invention is further described below in conjunction with preferred embodiments and accompanying drawings. Similar components in the accompanying drawings are represented by the same reference numerals. It should be understood by those skilled in the art that the following detailed description is illustrative rather than restrictive and should not be used to limit the scope of protection of the present invention.

[0049] It should be noted that, although the terms "first," "second," and "third" may be used to describe the embodiments of the present application, the description should not be limited to these terms. These terms are merely used to distinguish the descriptions. For example, without departing from the scope of the embodiments of the present application, "first" may also be referred to as "second," and similarly, "second" may also be referred to as "first."

[0050] See also Figure 1 As shown, it is a structural diagram of the purple sandware identification system based on image processing in this embodiment, the system includes:

[0051] The data acquisition module is used to collect images of purple clay pots and environmental data. The images of purple clay pots include grayscale images and HSV images of the purple clay pots. The environmental data include ambient light intensity and ambient humidity. In this embodiment, no specific limitation is imposed on the collection method of the images of purple clay pots and environmental data. Those skilled in the art can freely set it as long as the collection requirements of ambient light intensity and ambient humidity are met. Among them, the purple clay pot pictures can be taken under white natural light or fluorescent light, with the color temperature controlled at 5500K~6000K and a resolution higher than 1080p, and the image processing library OpenCV is used to read and convert them into grayscale images and HSV images. The environmental data can be collected by intelligent sensors when taking images.

[0052] Please continue reading Figure 1 As shown, the system includes:

[0053] The pre-processing module is connected to the data acquisition module, and is used for performing color normalization processing on the purple sandware image to obtain a corrected image.

[0054] Specifically, the pre-processing module sets the grayscale value of the i-th pixel after color normalization as LQi, and sets: LQi = (Ni / 255) γ ×255, γ is the correction factor, and Ni is the grayscale value of the i-th pixel point of the purple clay ware image before color normalization processing; the preprocessing module performs color normalization processing, and the preprocessing module can eliminate the color differences caused by shooting conditions, making subsequent image analysis more robust and ensuring that the analysis results are consistent with the actual color of the purple clay ware.

[0055] It is understandable that the setting of the correction factor is not specifically limited in this embodiment, and those skilled in the art can freely set it as long as the setting requirements of the correction factor are met. Among them, the optimal value of the correction factor is 2.2.

[0056] Please continue reading Figure 1 As shown, the system further includes:

[0057] The region division module is connected to the pre-processing module, and is used to calculate the gradient amplitude Pi of the i-th pixel point of the corrected image according to the gradient Tix in the horizontal direction and the gradient Tiy in the vertical direction of the i-th pixel point of the corrected image, and set

[0058] Pi=(Tix 2 +T iy 2 ) 0.5 ;

[0059] The region division module analyzes the relative position of the pixel point according to the gradient amplitude Pi of the i-th pixel point of the corrected image and the preset gradient amplitude threshold p0, wherein:

[0060] If Pi≤p0, the region division module determines that the pixel point is a non-edge pixel point; otherwise, the region division module determines that the pixel point is an edge pixel point;

[0061] The region division module uses the curve formed by connecting each edge pixel point as the outline of the purple clay ware, and uses the area within the outline of the purple clay ware as the purple clay ware area; the region division module calculates the gradient amplitude of each pixel in the image, can effectively distinguish the outline of the purple clay ware from the background, extract the key purple clay ware area for further analysis, and improve the accuracy of target recognition.

[0062] It can be understood that, in this embodiment, there is no specific limitation on the method for obtaining the gradient of the pixel point and the grayscale value of the pixel point. Those skilled in the art can set it freely, and it is only necessary to meet the requirements for obtaining the gradient of the pixel point and the grayscale value of the pixel point, wherein the gradient of the pixel point can be obtained by applying a 3x3 convolution kernel to the Sobel edge detection algorithm; the grayscale value of the pixel point can be read and converted by the OpenCV library in Python, and the grayscale value of the pixel point can be obtained; in this embodiment, there is no specific limitation on the setting of the preset gradient amplitude threshold, and those skilled in the art can set it freely, and it is only necessary to meet the setting requirements of the preset gradient amplitude threshold, wherein the optimal value of p0 is 0.3.

[0063] Please continue reading Figure 1 As shown, the system further includes:

[0064] A color feature analysis module is connected to the data acquisition module. The color feature analysis module is used to extract color features from the HSV image of the purple sandware and construct a color coefficient of the purple sandware area based on the extraction results, wherein:

[0065] The color feature analysis module sets the hue of the HSV image of the purple sandware as Hu, and the hue coefficient as SD, setting SD=1-|Hu-H0| / H0;

[0066] The color feature analysis module sets the saturation of the HSV image of the purple sandware to BH, and sets the saturation coefficient to BHD, setting BHD=1-|BH-BH0| / BH0;

[0067] The color feature analysis module sets the color coefficient of the purple sand ware as Y, setting Y=r1×SD+r2×BHD; the color feature analysis module can effectively capture the unique color expression of the purple sand ware by analyzing the color characteristics of the purple sand ware and constructing the color coefficient, thereby improving the accuracy of the identification of the purple sand ware clay material;

[0068] Among them, r1 is the hue weight, r2 is the saturation weight, r1+r2=1, H0 is the preset hue, BHO is the preset saturation, and when Y<0, the value of Y is 0.

[0069] Specifically, in this embodiment, there is no specific limitation on the method for obtaining the hue and saturation of the HSV image. Those skilled in the art can freely set it, as long as the requirements for obtaining the hue and saturation of the HSV image are met. Among them, it can be obtained through the built-in conversion function from RGB to HSV in OpenCV.

[0070] It can be understood that, in this embodiment, there is no specific limitation on the settings of each weight, preset hue and preset saturation. Those skilled in the art can set them freely as long as the setting requirements of each weight, preset hue and preset saturation are met. Among them, the optimal value of r1 is 0.7, the optimal value of r2 is 0.3, the optimal value of H0 is 10, and the optimal value of BH0 is 40%.

[0071] Please continue reading Figure 1 As shown, the system further includes:

[0072] A texture feature analysis module is connected to the region division module, and the texture feature analysis module is used to analyze the texture state of the purple sand ware region according to the contrast of the purple sand ware region and the image grayscale entropy of the purple sand ware region, and analyze the texture characteristics of the purple sand ware region according to the analysis results; it can be understood that in this embodiment, there is no specific limitation on the method for obtaining the contrast of the purple sand ware region and the image grayscale entropy of the purple sand ware region. Those skilled in the art can freely set it, as long as the requirements for obtaining the contrast of the purple sand ware region and the image grayscale entropy of the purple sand ware region are met, wherein it can be obtained through NumPy and OpenCV in Python.

[0073] See also Figure 2 As shown, the texture feature analysis module includes:

[0074] The contrast analysis unit is used to analyze the contrast state of the purple sand ware area according to the contrast d0 of the purple sand ware area, and construct a contrast coefficient based on the analysis result to quantify the contrast state of the purple sand ware area, wherein:

[0075] If d1≤d0≤d2, the contrast analysis unit determines that the contrast state of the purple sandware area is normal and sets the contrast coefficient to D1, setting D1=1; otherwise, the contrast analysis unit determines that the contrast state of the purple sandware area is abnormal. When d0<d1, the contrast analysis unit sets the contrast coefficient to D2, setting D2=(d1-d0) / △d; when d0>d2, the contrast analysis unit sets the contrast coefficient to D3, setting D3=(d0-d2) / △d; by evaluating the contrast state of the purple sandware area, the contrast analysis unit helps to analyze surface details, judge its production process and quality, and thus improve the accuracy of texture feature analysis;

[0076] Wherein, Δd is a preset contrast difference, Δd=(d1+d2) / 2, d1 is a first preset contrast, d2 is a second preset contrast, and d1<d2.

[0077] It is understandable that the present embodiment does not impose any specific restrictions on the setting of each preset contrast, and those skilled in the art can freely set it as long as the setting requirements of each preset contrast are met. Among them, the optimal value of d1 is 7, and the optimal value of d2 is 12.

[0078] Please continue reading Figure 2 As shown, the texture feature analysis module also includes:

[0079] A grayscale entropy analysis unit analyzes the grayscale entropy state of the purple sandware area according to the image grayscale entropy s0 and the preset grayscale entropy s1 of the purple sandware area, and constructs a grayscale entropy coefficient according to the analysis result to quantify the grayscale entropy state of the purple sandware area, wherein:

[0080] If s0≤s1, the grayscale entropy analysis unit determines that the grayscale entropy state of the purple sand ware area is abnormal, and sets the grayscale entropy coefficient to S1, setting S1=0; conversely, the grayscale entropy analysis unit determines that the grayscale entropy state of the purple sand ware area is normal, and sets the grayscale entropy coefficient to S2, setting S2=exp[3×(s0-s1) / (s0+s1)-3]; the grayscale entropy analysis unit can quantify the complexity of the texture through grayscale entropy analysis, providing quantitative support for the surface characteristics of the purple sand ware.

[0081] It is understandable that the setting of the preset grayscale entropy is not specifically limited in this embodiment, and those skilled in the art can freely set it as long as the setting requirements of the preset grayscale entropy are met. Among them, the optimal value of s1 is 1.8.

[0082] Please continue reading Figure 2 As shown, the texture feature analysis module also includes:

[0083] A texture analysis unit is connected to the contrast analysis unit and the grayscale entropy analysis unit, and the texture analysis unit analyzes the texture state of the purple sandware area according to the construction result of the contrast coefficient, and analyzes the texture characteristics of the purple sandware area according to the analysis result, wherein:

[0084] If x1×contrast coefficient+x2×grayscale entropy coefficient≤WL, the texture analysis unit determines that the texture state of the purple sand ware area is abnormal, and sets the texture feature of the purple sand ware area to WT1, setting WT1=0; otherwise, the texture analysis unit determines that the texture state of the purple sand ware area is normal, and sets the texture feature of the purple sand ware area to WT2, setting WT2=lg(x1×contrast coefficient+x2×grayscale entropy coefficient-WL+1) / lg2; the texture analysis unit determines the texture state of the purple sand ware by comprehensively analyzing the contrast and grayscale entropy, which helps to improve the scientificity and accuracy of the identification, thereby improving the accuracy of the analysis of the type of purple sand ware clay;

[0085] Among them, x1 is the contrast weight, x2 is the grayscale entropy weight, x1+x2=1, and WL is the preset texture coefficient.

[0086] It can be understood that in this implementation, there is no specific limitation on the setting of each weight and preset texture coefficient. Those skilled in the art can set it freely as long as the setting requirements of each weight and preset texture coefficient are met. Among them, the optimal value of x1 is 0.4, the optimal value of x2 is 0.6, and the optimal value of WL is 0.2.

[0087] Please continue reading Figure 1 As shown, the system further includes:

[0088] A gloss feature analysis module is connected to the texture feature analysis module. The gloss feature analysis module is used to classify pixel point types and analyze the gloss features of the purple sandware area based on the classification results.

[0089] See also Figure 3 As shown, the gloss feature analysis module includes:

[0090] The type analysis unit is used to classify the pixel type according to the grayscale value Lk of the k-th pixel in the purple sandware area and the preset grayscale value threshold Yz, where:

[0091] If Lk≤YZ, the type analysis unit determines that the pixel type is a Class I pixel; otherwise, the type analysis unit determines that the pixel type is a Class II pixel; by dividing pixels into different types, the type analysis unit can analyze different features and improve the effectiveness of gloss feature extraction.

[0092] Specifically, in this embodiment, there is no specific limitation on the setting of the preset grayscale value threshold, and those skilled in the art can set it freely, as long as the setting requirements of the preset grayscale threshold are met. Among them, the optimal value of YZ is 150.

[0093] Please continue reading Figure 3 As shown, the gloss feature analysis module also includes:

[0094] The gloss feature analysis unit is connected to the type analysis unit, and the gloss feature analysis unit calculates the average value PG of the grayscale value of the two types of pixels and sets LEm is the grayscale value of the mth second-class pixel, and M is the number of second-class pixels;

[0095] The gloss feature analysis unit analyzes the gloss coefficient GX of the purple sandware according to the average grayscale value PG of the two types of pixels and the average pixel value Pu of the purple sandware area, and sets GT=ln(PG / Pu) / ln2. Zn is the grayscale value of the nth pixel in the purple sandware area, and N is the number of pixels in the purple sandware area;

[0096] The gloss feature analysis unit analyzes the gloss feature of the purple clay ware area according to the gloss coefficient GX of the purple clay ware and the preset gloss coefficient GX0. If GX is less than GX0, the gloss feature analysis unit sets the gloss feature of the purple clay ware area to GT1, and sets GT1=(GX0-GX) / GX0. Otherwise, the gloss feature analysis unit sets the gloss feature of the purple clay ware area to GT2, and sets GT2=0. By analyzing the gloss coefficient, the gloss feature analysis unit can reveal the details of the surface treatment process of the purple clay ware, thereby improving the accuracy of the gloss feature analysis.

[0097] It is understandable that the present embodiment does not impose any specific limitation on the setting of the preset gloss coefficient, and those skilled in the art can freely set it as long as the setting requirements of the preset gloss coefficient are met. Among them, the optimal value of GX0 is 0.3.

[0098] Please continue reading Figure 1 As shown, the system further includes:

[0099] A judgment module is connected to the color feature analysis module and the gloss feature analysis module, and is used to analyze the type of purple clay material based on the construction results of the color coefficient of the purple clay area, the texture characteristics of the purple clay area, the gloss characteristics of the purple clay area, and the collected ambient light intensity and ambient humidity.

[0100] See also Figure 4 As shown, the judgment module includes:

[0101] The judgment unit is used to analyze the type of purple sandware clay material based on the construction result of the color coefficient of the purple sandware area, the texture characteristics of the purple sandware area, and the gloss characteristics of the purple sandware area, and output it to the user, wherein:

[0102] If w1×the color coefficient of the purple sand ware region+w2×the texture feature of the purple sand ware region+w3×the gloss feature of the purple sand ware region≤u0, the judgment unit determines that the purple sand ware clay is green clay; otherwise, the judgment unit determines that the purple sand ware clay is not green clay;

[0103] Among them, w1 is the color weight, w2 is the texture weight, w3 is the gloss weight, w1+w2+w3=1, and u0 is the preset judgment threshold; the judgment unit makes a scientific judgment on the type of purple clay material from a global perspective by combining the color coefficient, texture characteristics and gloss characteristics for comprehensive analysis, thereby improving the comprehensiveness and accuracy of the identification.

[0104] It can be understood that the present embodiment does not impose specific limitations on the settings of the weights and the preset judgment thresholds, and those skilled in the art can freely set them as long as the setting requirements of the weights and the preset judgment thresholds are met. Among them, the optimal value of w1 is 0.5, the optimal value of w2 is 0.3, the optimal value of w3 is 0.2, and the optimal value of u0 is 0.6.

[0105] Please continue reading Figure 4 As shown, the judgment module includes:

[0106] The light analysis unit is connected to the judgment unit, and is used to analyze the light state according to the collected ambient light intensity g0 and the preset light intensity g1, and process the analysis process of the purple sandware type according to the analysis results, wherein:

[0107] If g0≤g1, the illumination analysis unit determines that the illumination state is a normal state; otherwise, the illumination analysis unit determines that the illumination state is a strong light state;

[0108] The light analysis unit processes the analysis process of the purple clay ware type when the light state is a strong light state, and sets the preset judgment threshold after processing to u1, setting u1=u0×{1-β×ln[3×(g0-g1) / (g0+g1)+1] / ln5}, where β is a preset adjustment proportional coefficient; the light analysis unit can adjust the identification process of the purple clay ware according to the ambient light intensity, ensuring that the validity of the analysis results can be maintained under different lighting conditions, thereby improving the accuracy of the analysis of the clay type of the purple clay ware.

[0109] It can be understood that the present embodiment does not impose any specific restrictions on the setting of the preset light intensity and the preset adjustment ratio coefficient. Those skilled in the art can set them freely as long as they meet the setting requirements of the preset light intensity and the preset adjustment ratio coefficient. Among them, the optimal value of g1 is 800 Lux, and the optimal value of β is 0.36.

[0110] Please continue reading Figure 4 As shown, the judgment module also includes:

[0111] A humidity analysis unit is connected to the light analysis unit, and is used to analyze the humidity state according to the collected ambient humidity hs0 and the preset humidity hs1, and process the analysis process of the light state according to the analysis result, wherein:

[0112] If hs0≤hs1, the humidity analysis unit determines that the humidity state is a normal state; otherwise, the humidity analysis unit determines that the humidity state is a high humidity state;

[0113] The humidity analysis unit processes the analysis process of the illumination state when the humidity state is a high humidity state, sets the preset illumination intensity after processing to g2, and sets g2=g1×[1-η×(hs0-hs1) / hs1], where η is a preset correction proportional coefficient; the humidity unit can more accurately adjust the judgment result of the purple clay ware by evaluating the influence of the ambient humidity on the analysis process, ensuring its stability and reliability when the humidity changes, thereby improving the accuracy of the analysis of the clay type of the purple clay ware.

[0114] It can be understood that the present embodiment does not impose any specific limitation on the setting of the preset humidity and the preset correction proportional coefficient. Those skilled in the art can set them freely, as long as the setting requirements of the preset humidity and the preset correction proportional coefficient are met. Among them, the optimal value of hs1 is 70%, and the optimal value of η is 0.42.

[0115] See also Figure 5 As shown in FIG, it is a flow chart of the purple sandware identification method based on image processing in this embodiment, including:

[0116] Step S1, collecting purple sandware images and environmental data;

[0117] Step S2, performing color normalization processing on the purple sandware image to obtain a corrected image;

[0118] Step S3, dividing the corrected image into regions according to the horizontal gradient and the vertical gradient of each pixel point of the corrected image to obtain purple sandware regions;

[0119] Step S4, extracting color features from the HSV image of the purple sandware, and constructing a color coefficient of the purple sandware area based on the extraction results;

[0120] Step S5, analyzing the texture state of the purple sand ware region according to the contrast of the purple sand ware region and the image grayscale entropy of the purple sand ware region, and analyzing the texture characteristics of the purple sand ware region according to the analysis results;

[0121] Step S6, classifying the pixel types and analyzing the gloss characteristics of the purple sandware area based on the classification results;

[0122] Step S7: analyzing the type of purple sand ware clay material based on the color coefficient construction results, texture characteristics, gloss characteristics, and collected ambient light intensity and ambient humidity of the purple sand ware area.

[0123] Specifically, the image processing-based purple sandware identification method and system described in this embodiment is applied to identify the type of purple sandware clay. By collecting images of the purple sandware and environmental data, color normalization and contour extraction are performed. Through a comprehensive analysis of color coefficients, texture state, and gloss characteristics, combined with ambient light and humidity, the clay type of the purple sandware can be accurately determined. This solution significantly improves the efficiency and accuracy of purple sandware identification through a scientific and standardized approach.

[0124] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not limitations on the implementation methods of the present invention. For ordinary technicians in this field, other different forms of changes or modifications can be made based on the above description. It is impossible to list all the implementation methods here. All obvious changes or modifications derived from the technical solution of the present invention are still within the scope of protection of the present invention.

Claims

1. A purple sandware identification system based on image processing, characterized in that: include: Data acquisition module, used to collect images of purple clay pots and environmental data; A pre-processing module is used to perform color normalization on the purple sandware image to obtain a corrected image; A region division module is used to divide the corrected image into regions according to the horizontal gradient and the vertical gradient of each pixel point of the corrected image to obtain the purple sandware region; The color feature analysis module is used to extract the color features of the HSV image of the purple sandware and construct the color coefficient of the purple sandware area based on the extraction results; A texture feature analysis module is used to analyze the texture state of the purple sand ware area according to the contrast of the purple sand ware area and the image grayscale entropy of the purple sand ware area, and analyze the texture features of the purple sand ware area according to the analysis results; The gloss feature analysis module is used to classify pixel types and analyze the gloss features of the purple sandware area based on the classification results; The judgment module is used to analyze the type of purple clay material based on the construction results of the color coefficient of the purple clay area, texture characteristics, gloss characteristics and the collected ambient light intensity and ambient humidity.

2. The purple sandware identification system based on image processing according to claim 1 is characterized in that: The region division module calculates the gradient amplitude Pi of the i-th pixel point of the corrected image according to the gradient Tix in the horizontal direction and the gradient Tiy in the vertical direction of the i-th pixel point of the corrected image, and analyzes the relative position of the pixel point according to the gradient amplitude Pi of the i-th pixel point of the corrected image and the preset gradient amplitude threshold p0, and the relative position of the pixel point includes non-edge pixels and edge pixels; The region division module uses a curve formed by connecting the edge pixel points as the outline of the purple sand ware, and uses the area within the outline of the purple sand ware as the purple sand ware area.

3. The purple sandware identification system based on image processing according to claim 2 is characterized in that: The color feature analysis module sets the hue of the HSV image of the purple sandware as Hu and the hue coefficient as SD; The color feature analysis module sets the saturation of the HSV image of the purple sandware to BH and the saturation coefficient to BHD; The color feature analysis module sets the color coefficient of the purple sandware as Y, setting Y=r1×SD+r2×BHD; Among them, r1 is the hue weight and r2 is the saturation weight.

4. The purple sandware identification system based on image processing according to claim 3 is characterized in that: The texture feature analysis module includes a contrast analysis unit, which is used to analyze the contrast state of the purple sandware area according to the contrast d0 of the purple sandware area, and construct a contrast coefficient based on the analysis result, and the contrast coefficient includes D1, D2 and D3; The texture feature analysis module also includes a grayscale entropy analysis unit, which is used to analyze the grayscale entropy state of the purple sand ware area based on the image grayscale entropy s0 and the preset grayscale entropy s1 of the purple sand ware area. The grayscale entropy state of the purple sand ware area includes a normal state and an abnormal state. If the grayscale entropy state of the purple sand ware area is an abnormal state, the grayscale entropy coefficient is set to S1. If the grayscale entropy state of the purple sand ware area is a normal state, the grayscale entropy coefficient is set to S2.

5. The purple sandware identification system based on image processing according to claim 4 is characterized in that: The texture feature analysis module also includes a texture analysis unit, which is used to analyze the texture state of the purple sand ware area according to the construction result of the contrast coefficient. The texture state of the purple sand ware area includes a normal state and an abnormal state. If the texture state of the purple sand ware area is an abnormal state, the texture feature of the purple sand ware area is set to WT1. If the texture state of the purple sand ware area is a normal state, the texture feature of the purple sand ware area is set to WT2.

6. The purple sandware identification system based on image processing according to claim 5 is characterized in that: The gloss feature analysis module includes a type analysis unit, which is used to classify pixel types according to the grayscale value Lk of the kth pixel in the purple sandware area and the preset grayscale value threshold YZ. The pixel types include type I pixels and type II pixels. The gloss feature analysis module further includes a gloss feature analysis unit, which is used to calculate the average grayscale value PG of the two types of pixel points, and analyze the gloss coefficient GX of the purple sandware according to the average grayscale value PG of the two types of pixel points and the average value Pu of the pixel points in the purple sandware area; The gloss feature analysis unit analyzes the gloss feature of the purple clay ware area according to the gloss coefficient GX of the purple clay ware and the preset gloss coefficient GX0. If GX<GX0, the gloss feature analysis unit sets the gloss feature of the purple clay ware area to GT1; otherwise, the gloss feature analysis unit sets the gloss feature of the purple clay ware area to GT2.

7. The purple sandware identification system based on image processing according to claim 6 is characterized in that: The judgment module includes a judgment unit, which is used to analyze the type of purple sandware clay according to the construction result of the color coefficient of the purple sandware area, the texture characteristics of the purple sandware area, and the gloss characteristics of the purple sandware area, wherein: If w1×the color coefficient of the purple sand ware region+w2×the texture feature of the purple sand ware region+w3×the gloss feature of the purple sand ware region≤u0, the judgment unit determines that the purple sand ware clay is green clay; otherwise, the judgment unit determines that the purple sand ware clay is not green clay; Among them, w1 is the color weight, w2 is the texture weight, w3 is the gloss weight, and u0 is the preset judgment threshold.

8. The purple sandware identification system based on image processing according to claim 7 is characterized in that: The judgment module includes a light analysis unit, which is used to analyze the light state according to the collected ambient light intensity g0 and the preset light intensity g1. The light state includes a normal state and a strong light state; The illumination analysis unit processes the analysis process of the purple clay ware type when the illumination state is a strong light state, and sets the preset judgment threshold after processing as u1.

9. The purple sandware identification system based on image processing according to claim 8 is characterized in that: The judgment module further includes a humidity analysis unit, which is used to analyze the humidity state according to the collected ambient humidity hs0 and the preset humidity hs1, and the humidity state includes a normal state and a high humidity state; The humidity analysis unit processes the analysis process of the illumination state when the humidity state is a high humidity state, and sets the preset illumination intensity after processing to g2.

10. A method for identifying purple clay wares based on image processing, applied to the purple clay wares identification system based on image processing according to any one of claims 1 to 9, characterized in that: include: Step S1, collecting purple sandware images and environmental data; Step S2, performing color normalization processing on the purple sandware image to obtain a corrected image; Step S3, dividing the corrected image into regions according to the horizontal gradient and the vertical gradient of each pixel point of the corrected image to obtain purple sandware regions; Step S4, extracting color features from the HSV image of the purple sandware, and constructing a color coefficient of the purple sandware area based on the extraction results; Step S5, analyzing the texture state of the purple sand ware region according to the contrast of the purple sand ware region and the image grayscale entropy of the purple sand ware region, and analyzing the texture characteristics of the purple sand ware region according to the analysis results; Step S6, classifying the pixel types and analyzing the gloss characteristics of the purple sandware area based on the classification results; Step S7: analyzing the type of purple sand ware clay material based on the color coefficient construction results, texture characteristics, gloss characteristics, and collected ambient light intensity and ambient humidity of the purple sand ware area.

Images