CN-121982701-A - Jadeite class identification method, program product, electronic equipment and storage medium

CN121982701ACN 121982701 ACN121982701 ACN 121982701ACN-121982701-A

Abstract

The application provides a jadeite class identification method, a program product, electronic equipment and a storage medium, wherein the method comprises the steps of collecting color data of jadeite to be identified; the color data comprises at least one of color space parameters, tristimulus values and chromaticity coordinates, the color data is processed to obtain color feature vectors of jades to be identified, the color feature vectors are input into a trained jades classification model to obtain jades classification identification results, and the jades classification model is obtained by training a preset machine learning classification model through sample color feature vectors corresponding to jades samples of different classes. The jadeite to be identified is converted into color space parameters, tristimulus values and chromaticity coordinates, so that color observation which is originally dependent on personal experience is converted into digital information, and errors and inconsistent results caused by factors such as experience, light and the like in manual identification are reduced. The whole identification process only needs to carry out optical measurement on the surface of the jadeite, and destructive operations such as cutting and the like are not needed, so that nondestructive detection is realized.

Inventors

WU JINLIN
HAN CUICUI
JIANG QINGYANG
LI XUYANG
ZHU MEIDONG
HUANG XIAOQIANG
LIU HAI

Assignees

睹煜(上海)信息服务有限公司

Dates

Publication Date: 20260505
Application Date: 20260127

Claims (12)

1. A method for identifying a class of jadeite, comprising: Collecting color data of jadeite to be identified, wherein the color data comprises at least one of color space parameters, tristimulus values and chromaticity coordinates; Processing the color data to obtain a color feature vector of the jadeite to be identified; And inputting the color feature vectors into a trained jadeite classification model to obtain a jadeite class identification result, wherein the jadeite classification model is obtained by training a preset machine learning classification model through sample color feature vectors corresponding to jadeite samples of different classes.
2. The method according to claim 1, wherein collecting color data of the jadeite to be identified comprises: Collecting spectral reflectivity or color signals of the surface of the jadeite to be identified by using a color measuring instrument under a preset light source and/or a preset observation condition; The spectral reflectance or color signal is converted to at least one of the color space parameters, tristimulus values, and chromaticity coordinates based on a colorimetry conversion model.
3. The method of claim 1, wherein the color space parameters include a luminance value, a red-green color value, and a yellow-blue color value in a CIELab color system, wherein the tristimulus values are quantization indexes for describing three primary color stimulus levels of human retina for color perception, and wherein the tristimulus values include a red primary color, a green primary color, and a blue primary color; processing the color data to obtain a color feature vector of the jadeite to be identified, wherein the color feature vector comprises the following components: Normalizing or standardizing at least one of the color space parameters, the tristimulus values and the chromaticity coordinates to obtain preprocessed data; and combining the preprocessed data according to a preset sequence to obtain the color feature vector.
4. A method according to claim 3, wherein processing the color data to obtain a color feature vector of the jadeite to be identified comprises: calculating color saturation and/or hue angle based on the red-green and yellow-blue color values in the color space parameters, and taking the color saturation and/or hue angle as a derived color feature; Inputting the color feature vector into a trained jadeite classification model to obtain a jadeite class identification result, wherein the method comprises the following steps of: And the color feature vector and the derived color feature are input into a trained jadeite classification model together to obtain a jadeite class identification result, and the jadeite classification model is obtained by training a preset machine learning classification model through sample color feature vectors and sample derived color features corresponding to jadeite samples of different classes.
5. The method of claim 1, wherein prior to inputting the color feature vector into the trained jadeite classification model, the method further comprises: acquiring a plurality of classes of jadeite samples, and acquiring sample color data of the jadeite samples; Constructing a sample color feature vector of the jadeite sample based on sample color data of the jadeite sample; The method comprises the steps of taking the class of a jadeite sample as a label, forming a training data set by sample color feature vectors corresponding to the jadeite sample, inputting the training data set into a preset machine learning classification model for training, obtaining the trained jadeite classification model after model verification of the trained jadeite classification model, updating key parameters of the machine learning classification model by adopting a cross verification method in the training process, wherein the preset machine learning classification model comprises a support vector machine model, and the key parameters comprise penalty factors and kernel function parameters.
6. The method of claim 1, wherein the trained jadeite classification model forms a classification decision boundary that distinguishes between different jadeite classes, the classification decision boundary defining a cluster distribution of sample color feature vectors corresponding to different classes of jadeite samples in a feature space; Inputting the color feature vector into a trained jadeite classification model to obtain a jadeite class identification result, wherein the method comprises the following steps of: and inputting the color feature vector into a trained jadeite classification model, wherein the jadeite classification model determines the class of the jadeite to be identified based on the classification decision boundary, and generates the jadeite class identification result.
7. The method of claim 6, wherein the class of jadeite samples comprises a first regional jadeite and a second regional jadeite; the classification decision boundary is used for dividing a first color cluster corresponding to the jadeite in a first area and a second color cluster corresponding to the jadeite in a second area, wherein in a characteristic subspace of the color space parameter, the statistic value of the characteristic vector of the first color cluster in a brightness value and a red-green product value is higher than that of the second color cluster, or the yellow-blue product value of the first color cluster is lower than that of the second color cluster; In the feature subspace of the tristimulus values, the statistical value of the color feature vector of the first color cluster is higher than that of the second color cluster; in the feature subspace of the chromaticity coordinates, the distribution concentration of the color feature vectors of the first color cluster is higher than that of the second color cluster.
8. The method according to claim 1, wherein inputting the color feature vector into a trained jadeite classification model to obtain a jadeite class discrimination result comprises: Collecting surface optical characteristic data of the jadeite to be identified, wherein the surface optical characteristic data comprise glossiness parameters and/or transparency indexes, the glossiness parameters are obtained by measuring specular reflection light intensity of the surface of the jadeite through a glossiness measuring instrument, and the transparency indexes are obtained by measuring transmitted light of the jadeite and calculating the intensity ratio of incident light to transmitted light; acquiring an optical feature vector of the jadeite to be identified based on the surface optical characteristic data; Performing fusion processing on the color feature vector and the optical feature vector to generate a fused feature; inputting the integrated features into a trained jadeite classification model to obtain a jadeite class identification result, wherein the jadeite classification model is obtained by training a preset machine learning classification model through sample fusion features corresponding to jadeite samples of different classes.
9. The method according to claim 1, wherein the method further comprises: Acquiring time sequence color data of the jadeite to be identified according to time sequence under a plurality of preset measurement conditions; Calculating at least one dynamic stability parameter reflecting the dynamic characteristics of the jadeite color to be identified based on the time sequence color data; Fusing the dynamic stability parameters with the color feature vectors to generate fusion feature vectors; Inputting the color feature vector into a trained jadeite classification model to obtain a jadeite class identification result, wherein the method comprises the following steps of: Inputting the fusion feature vector into a trained jadeite classification model to obtain a jadeite class identification result, wherein the jadeite classification model is obtained by training a preset machine learning classification model through sample fusion feature vectors corresponding to jadeite samples of different classes.
10. A computer program product comprising computer program instructions which, when executed by a processor, perform the method of any of claims 1 to 9.
11. An electronic device comprising a processor and a memory, the memory storing computer program instructions that, when executed by the processor, perform the method of any one of claims 1 to 9.
12. A computer readable storage medium, characterized in that it has stored thereon computer program instructions which, when executed by a processor, perform the method according to any of claims 1 to 9.

Description

Jadeite class identification method, program product, electronic equipment and storage medium Technical Field The application relates to the technical field of jadeite identification, in particular to a jadeite type identification method, a program product, electronic equipment and a storage medium. Background The prior jadeite identification method mainly has two types, namely, the first identification mode relies on subjective experience of an identifier to observe the characteristics of color, luster and the like. This approach lacks objective criteria, results are unstable and less efficient. The second mode of identification is determined by detecting trace elements. This approach typically requires micro-destructive or destructive sampling, which is difficult to identify without loss. Disclosure of Invention An object of an embodiment of the present application is to provide a jadeite class identification method, a program product, an electronic device, and a storage medium for improving the above-described problems. In a first aspect, an embodiment of the present application provides a method for identifying a jadeite class, including collecting color data of a jadeite to be identified, where the color data includes at least one of a color space parameter, a tristimulus value, and a chromaticity coordinate, processing the color data to obtain a color feature vector of the jadeite to be identified, inputting the color feature vector into a trained jadeite classification model to obtain a jadeite class identification result, and training a preset machine learning classification model by the jadeite classification model through sample color feature vectors corresponding to different types of jadeite samples. In the implementation process, the color data of the jades are collected first, the jades to be identified are converted into a set of objective color space parameters, tristimulus values and chromaticity coordinates, so that subjective color observation which originally depends on personal experience is converted into digital information, and subjective errors and inconsistent results caused by factors such as experience, light and the like in manual identification are reduced. And then, through standardized processing and eigenvector construction of the color data, a unified and clean data format is provided for subsequent computer analysis, so that data measured by different batches and different instruments can be put together for fair comparison and analysis. Finally, the processed color feature vectors are input into a pre-trained jadeite classification model, so that the rapid discrimination of the place of origin of the jadeite is realized. The whole identification process only needs to carry out optical measurement on the surface of the jadeite, and destructive operations such as cutting and the like are not needed, so that nondestructive detection is realized. The method reduces the dependence on the qualification specialists, and provides a jadeite identification scheme which has higher efficiency and better consistency and does not damage samples. Optionally, in the embodiment of the application, collecting the color data of the jadeite to be identified comprises collecting the spectral reflectivity or color signal of the surface of the jadeite to be identified by using a color measuring instrument under the preset light source and/or the preset observation condition, and converting the spectral reflectivity or color signal into at least one of color space parameters, tristimulus values and chromaticity coordinates based on a colorimetry conversion model. In the implementation process, the color measuring instrument is used for collecting the spectral reflectance data of the jadeite, so that subjective observation of human eyes under uncertain ambient light is reduced, the acquisition source of the color information becomes objective, and the color information is not interfered by human experience or environment. The original optical data are accurately calculated into a series of recognized colorimetry parameters such as color space parameters, tristimulus values, chromaticity coordinates and the like through a colorimetry conversion model, so that the digitization and standardization of color characteristics are realized. Meanwhile, the whole data acquisition and conversion process only needs to carry out optical irradiation on the jadeite, and no physical contact or sample damage is needed, so that nondestructive detection is realized. Optionally, in the embodiment of the application, the color space parameters comprise a brightness value, a red green quality value and a yellow blue quality value in a CIELab color system, tristimulus values are used for describing quantization indexes of three primary color stimulus degrees of human retina on color perception, the tristimulus values comprise a red primary color, a green primary color and a blue primary color, chromaticity coordinates