Reconstituted amber identification method and image acquisition method

Abstract

The invention provides a reconstructed amber identification method and an image acquisition method, and the method comprises the steps: obtaining an amber morphological feature image of an amber to be detected; matching the amber morphological feature image with a sample feature image in a map database to obtain a matching degree between the amber morphological feature image and the sample feature image; according to the matching degree, obtaining an atlas identification result of the amber to be detected; wherein the spectrum identification result is used for representing whether the amber to be detected is natural amber. According to the scheme, amber identification is realized in a mode of comparing and matching the atlas database, so that the amber identification difficulty is reduced, and the amber identification efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to a method for identifying reconstructed amber and an image acquisition method. Background Art

[0002] Amber is a resin secreted by plants such as Pinaceae and Rosales. It is buried underground by geological action, completes molecular polymerization, evaporates unsaturated volatile components more completely, and is a petrochemical resin with strong physical and chemical stability. Since natural amber fragments, powders, and scraps cannot be processed into jewelry, natural amber fragments or powders are used as raw materials, pressurized and heated to a certain degree in a pressure furnace to melt and sinter, and then form larger pieces of reconstructed amber. Organic substances such as dyes, flavors, and adhesives can also be added during the pressing process to make the amber more valuable. The artificial amber formed by this process is called reconstructed amber.

[0003] At present, the detection of reconstructed amber mostly uses density measurement, refractive index measurement, fluorescence observation, magnification inspection, infrared spectrum measurement, laser Raman spectrum measurement, etc. However, these detection methods are limited to laboratory scenes and detection methods, and the detection efficiency of amber is low. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a method for identifying reconstructed amber and an image acquisition method, so as to improve the detection efficiency of reconstructed amber.

[0005] In a first aspect, an embodiment of the present application provides a method for identifying reconstructed amber, the method comprising: obtaining an amber morphological characteristic image of amber to be tested; matching the amber morphological characteristic image with a sample characteristic image in a graph database to obtain a matching degree between the amber morphological characteristic image and the sample characteristic image; based on the matching degree, obtaining a graph identification result of the amber to be tested; wherein the graph identification result is used to characterize whether the amber to be tested is natural amber.

[0006] In the process of implementing the above scheme, on the one hand, the above scheme realizes amber identification through the method of comparison and matching of the atlas database, which is conducive to reducing the difficulty of amber identification and thus improving the efficiency of amber identification; on the other hand, the above scheme realizes amber identification through image matching, which is conducive to reducing the identification damage of amber and thus realizing lossless or low-loss identification of amber.

[0007] In an implementation of the first aspect, obtaining the amber morphological characteristic image of the amber to be tested includes: obtaining at least one of a surface magnified characteristic image, a light-transmitting structure characteristic image, and a fluorescence characteristic image of the amber to be tested.

[0008] In the process of implementing the above scheme, reconstructed amber can be identified by at least one of the surface magnification characteristics, light-transmitting structure characteristics and fluorescence characteristics. On the one hand, the above-mentioned reconstructed amber identification method can be applied to more application scenarios, which is beneficial to improving the adaptability of the above-mentioned reconstructed amber identification method. On the other hand, since there are certain differences between reconstructed amber and natural amber in surface magnification characteristics, light-transmitting structure characteristics and fluorescence characteristics, identifying reconstructed amber by at least one of the surface magnification characteristics, light-transmitting structure characteristics and fluorescence characteristics is beneficial to improving the identification accuracy of the above-mentioned reconstructed amber identification method.

[0009] In an implementation of the first aspect, matching the amber morphological characteristic image with a sample characteristic image in an atlas database to obtain a degree of matching between the amber morphological characteristic image and the sample characteristic image includes: obtaining a first feature descriptor of the amber morphological characteristic image; wherein the first feature descriptor is used to characterize image features of the amber morphological characteristic image; matching the first feature descriptor with a second feature descriptor of the sample characteristic image in the atlas database to obtain a degree of matching between the first feature descriptor and the second feature descriptor; wherein the second feature descriptor is used to characterize image features of the sample characteristic image.

[0010] In the implementation process of the above scheme, on the one hand, the feature descriptors can capture key information in the image such as texture, shape, color, etc., and the matching between feature descriptors is helpful to improve the identification accuracy of the above-mentioned reconstructed amber identification method; on the other hand, feature descriptors usually have a lower dimension, which is helpful to reduce the amount of calculation during image matching, thereby helping to improve the identification efficiency of the above-mentioned reconstructed amber identification method.

[0011] In an implementation of the first aspect, a multi-scale amber morphological feature image of the amber to be tested is obtained;

[0012] The matching of the amber morphological feature image with the sample feature image in the atlas database to obtain the matching degree between the amber morphological feature image and the sample feature image includes: obtaining a first feature descriptor of the amber morphological feature image at multiple scales; fusing the first feature descriptors at multiple scales to obtain a third feature descriptor; matching the third feature descriptor with the second feature descriptor of the sample feature image in the atlas database to obtain the matching degree between the third feature descriptor and the second feature descriptor.

[0013] In the implementation process of the above scheme, by fusing the multi-scale first feature descriptors, a third feature descriptor containing multi-scale information is obtained, and the third feature descriptor is used for image matching, which is beneficial to improve the matching accuracy, thereby improving the identification accuracy of the above-mentioned reconstructed amber identification method.

[0014] In an implementation of the first aspect, the method further includes: when the spectrum identification result is reconstructed amber, determining the reconstructed amber type of the amber to be tested according to the matching degree.

[0015] In the process of implementing the above scheme, it is possible to determine whether the amber to be tested is natural amber, and also to determine the type of amber to be tested when it is reconstructed amber, so that the above reconstructed amber identification method can be applied to more application scenarios, which is conducive to improving the adaptability of the above reconstructed amber identification method.

[0016] In an implementation of the first aspect, after obtaining the amber morphological characteristic image of the amber to be tested, the method further includes: performing at least one of denoising, contrast enhancement, size standardization and edge enhancement processing on the amber morphological characteristic image.

[0017] In the process of implementing the above scheme, the matching accuracy of the image is improved by performing at least one of denoising, contrast enhancement, size standardization and edge enhancement processing on the amber morphological feature image, thereby improving the identification accuracy of the above-mentioned reconstructed amber identification method.

[0018] In a second aspect, an embodiment of the present application provides an image acquisition method, which comprises: extracting sample morphological features from the morphological image of the amber sample, and obtaining a sample morphological feature image; wherein the sample morphological features comprise: at least one of sample surface magnification features, sample light-transmitting structure features, and sample fluorescence features; and the sample morphological feature image is used to construct a map database.

[0019] In an implementation of the second aspect, the method further includes: acquiring a sample morphological feature descriptor of the sample morphological feature image; wherein the sample morphological feature descriptor is used to construct the atlas database.

[0020] In an implementation of the second aspect, extracting the sample morphological features from the morphological image of the amber sample includes: extracting the sample morphological features from the morphological images of natural amber samples and / or reconstructed amber samples.

[0021] In an implementation of the second aspect, the sample morphological features are extracted from the morphological image of the reconstructed amber sample, including: extracting the sample morphological features from the morphological image of at least one of the reconstructed amber samples selected from the group consisting of reconstructed Russian white flower honey, reconstructed full wax white flower golden twisted honey, reconstructed golden twisted honey, reconstructed half mountain and half water flower honey, reconstructed chicken oil yellow beeswax, reconstructed roasted old beeswax, imitation blood beeswax, imitation Burmese flower amber, and imitation Fushun root amber.

[0022] In a third aspect, an embodiment of the present application provides a reconstructed amber identification device, the device comprising:

[0023] An amber morphological characteristic image acquisition module is used to acquire an amber morphological characteristic image of the amber to be tested;

[0024] A matching degree acquisition module, used for matching the amber morphological characteristic image with the sample characteristic image in the atlas database to obtain the matching degree between the amber morphological characteristic image and the sample characteristic image;

[0025] The spectrum identification result acquisition module is used to obtain the spectrum identification result of the amber to be tested according to the matching degree; wherein the spectrum identification result is used to indicate whether the amber to be tested is natural amber.

[0026] In a fourth aspect, an embodiment of the present application provides an electronic device, comprising: a processor, a memory, and a communication bus, wherein the processor and the memory communicate with each other through the communication bus; the memory stores computer program instructions that can be executed by the processor, and when the computer program instructions are read and run by the processor, the method provided by the first aspect or any possible implementation of the first aspect or the second aspect or any possible implementation of the second aspect is executed.

[0027] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which computer program instructions are stored. When the computer program instructions are read and executed by a processor, the method provided by the first aspect or any possible implementation of the first aspect or the second aspect or any possible implementation of the second aspect is executed. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0029] Figure 1A schematic diagram of a process for identifying reconstructed amber provided in an embodiment of the present application;

[0030] Figure 2 A schematic diagram of an amber sample provided in an embodiment of the present application;

[0031] Figure 3 The amber sample image and its corresponding sample characteristic image provided in the embodiment of the present application;

[0032] Figure 4 A schematic diagram of the structure of a reconstructed amber identification device provided in an embodiment of the present application;

[0033] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] The following will describe the technical solutions in the embodiments of the present application in conjunction with the accompanying drawings in the embodiments of the present application. The following embodiments are only used to more clearly illustrate the technical solutions of the present application, and are therefore only used as examples, and cannot be used to limit the scope of protection of the present application.

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of the present application; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" in the specification and claims of the present application and the above-mentioned figure descriptions and any variations thereof are intended to cover non-exclusive inclusions. In the description of the embodiments of the present application, the technical terms "first" and "second" are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is more than two, unless otherwise clearly and specifically defined. The reference to "embodiment" in this article means that the specific features, structures or characteristics described in conjunction with the embodiment may be included in at least one embodiment of the present application. The appearance of this phrase in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments. In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0036] Most of the regeneration processes used in regenerated amber (also known as pressed amber, melted amber, fused amber, molded amber, and second-generation amber) are achieved through amber pressure furnaces. The regeneration process is carried out in a sealed furnace cabin. Technicians control the pressure by adjusting the amount of gas entering the pressure furnace and control the temperature by setting the intelligent temperature control system. Ultimately, under the combined effects of temperature, pressure, time, and ambient gases, various regeneration processes such as purification, color baking, aroma enhancement, and explosion of amber are achieved.

[0037] At present, most of the related technologies use density measurement, refractive index measurement, fluorescence observation, magnification inspection, infrared spectrum measurement, laser Raman spectrum measurement, etc. However, they are limited to laboratory scenarios and detection methods. The above detection schemes rely on manual operation, which takes a long time and has high detection costs, and the overall detection efficiency is low.

[0038] Based on this, an embodiment of the present application provides a method for identifying reconstructed amber. The method matches the amber morphological characteristic image of the amber to be tested with the sample characteristic image in the atlas database, thereby obtaining the matching degree between the amber morphological characteristic image and the sample characteristic image, and then determines the identification result of whether the amber to be tested is natural amber through the matching degree. On the one hand, the above scheme realizes amber identification through atlas database comparison and matching, which is conducive to reducing the difficulty of amber identification; on the other hand, the above scheme realizes amber identification through image matching, which is conducive to reducing the identification damage of amber, thereby realizing lossless or low-loss identification of amber.

[0039] The following is a detailed introduction to the identification method of reconstructed amber:

[0040] See also Figure 1 The present invention provides a method for identifying reconstructed amber, the method comprising:

[0041] Step S110: obtaining an amber morphological characteristic image of the amber to be tested;

[0042] Step S120: matching the amber morphological characteristic image with the sample characteristic image in the atlas database to obtain the matching degree between the amber morphological characteristic image and the sample characteristic image;

[0043] Step S130: Obtain the spectrum identification result of the amber to be tested according to the matching degree; wherein the spectrum identification result is used to characterize whether the amber to be tested is natural amber.

[0044] It is understood that the amber morphological characteristic image in the embodiment of the present application refers to an image that can distinguish natural amber from reconstructed amber, and the image contains specific features of amber to provide a visual basis for amber identification. The above-mentioned specific features may include features of the texture, structure, etc. of amber. An optional implementation of the above-mentioned amber morphological characteristic image is described below:

[0045] As an optional implementation of the above-mentioned reconstructed amber identification method, the above-mentioned step S110 includes: obtaining at least one of the surface magnified characteristic image, the light-transmitting structure characteristic image and the fluorescence characteristic image of the amber to be tested.

[0046] The above-mentioned surface magnification characteristics, light-transmitting structure characteristics and fluorescence characteristics are introduced below respectively:

[0047] (1) Surface magnification features: The outer surface of reconstructed amber usually has stirring patterns such as leaf vein-like patterns, loofah-like patterns, and irregular strip-like patterns. The stirring patterns are caused by the interweaving of the pressed and unpressed parts of the amber powder or fragments during the pressing process, resulting in stirring patterns on the outer surface of the reconstructed amber. During the pressing process, due to the effects of heat and pressure, the amber powder or fragments will partially melt and recombine to form larger clumps. In this process, due to the uneven distribution of the powder or fragments and the different degrees of melting, a special flow pattern structure will be formed in the reconstructed amber. The flow patterns of natural amber are usually layers of texture formed by the secretion of resins and their flow on the surface. It takes a long time to form, so the overall shape is round and natural. The stirring patterns of reconstructed amber are unique structures formed during a single pressing process, which are rigid and stiff.

[0048] In the process of making reconstructed amber, the steps of melting and re-cooling and pressing will cause stirring marks to appear on the outer surface. Under the principles of fluid mechanics, amber powder or small pieces are re-melted and combined under high temperature and pressure. Due to the fluidity of the molten material, they are stirred and pressed inside the mold, forming a texture with a certain directionality, showing irregular flow patterns in the shape of leaf veins, loofahs or strips, which are significantly different from the flow patterns formed by the natural dripping of resin in natural amber. The flow patterns of natural amber are formed by the gradual accumulation and drying of resin over a long period of time. They usually show natural and irregular flows, while the stirring patterns of pressed amber show more regular and directional texture characteristics due to artificial stirring and pressure. Therefore, reconstructed amber generally has this surface characteristic texture. The difference between the surface textures can be used to determine whether the amber is natural.

[0049] (2) Translucent structural characteristics: Reconstructed amber generally has internal structures such as broken spots, broken particles and broken powder. This is mainly caused by the special conditions and methods in the manufacturing process of reconstructed amber. It is the explicit communication of the physical and chemical changes in the production process of reconstructed amber. Among them:

[0050] Broken porphyry structure: In the production process of reconstructed amber, when amber particles are broken and recombined under high temperature and pressure, some larger particles are retained, forming a structure similar to porphyry. When this structure is observed under a microscope, irregular larger particles can be seen surrounded by fine particles or powder.

[0051] Crushed structure: During the production of reconstructed amber, the raw amber is broken into small particles and then recombined under high temperature and pressure. These particles may not completely melt during the combining process, resulting in a crushed structure in the final product.

[0052] Broken powder structure: This structure may be due to the uneven distribution of temperature and pressure during the pressing process of the amber powder, which leads to the failure of complete fusion of the powder, thus forming a powder-like structure in the reconstructed amber.

[0053] There are certain differences between the internal structural characteristics of reconstructed amber and that of natural amber. The internal structural characteristics of amber can be more easily obtained after light is transmitted through it. Therefore, the light-transmitting structural characteristics of amber can be used to determine whether it is natural amber.

[0054] (3) Fluorescence characteristics: The fluorescence reaction morphological characteristics of reconstructed amber, such as blood streak characteristics, meat jelly characteristics, splicing characteristics, convex pattern characteristics, and dot pattern characteristics, are mainly caused by the inconsistent fluorescence characteristics of different raw materials during the production of reconstructed amber, the uneven distribution of temperature and pressure during the pressing process, and the possible addition of fluorescent dyes. The following is a detailed description of these characteristics and the formation principle:

[0055] Blood streaks: These are usually red or brown streaks inside reconstructed amber, similar to the structure of capillaries. They are caused by uneven melting and fusion of amber powder or fragments during the pressing process, and excessive epoxy resin content in the colorant, which appears due to differences in fluorescent properties in heterogeneous polymers.

[0056] Meat skin jelly feature: This feature is manifested as irregular block or mass fluorescence inside the reconstructed amber, similar to the appearance of meat skin jelly. This is caused by the incomplete fusion of the boundaries between different amber particles during the pressing process, or by the special reflection and scattering of internal bubbles and voids under fluorescence.

[0057] Synthesis characteristics: Under fluorescence, different parts of reconstructed amber may show different fluorescence reactions, forming a puzzle-like pattern. This is because reconstructed amber is composed of amber fragments from multiple different batches or different sources, and the fluorescence characteristics of each part may be different.

[0058] Raised pattern features: This feature appears as irregular raised fluorescence inside the reconstructed amber, which may be more obvious under fluorescence. This may be caused by the uneven local compression of the amber material during the pressing process.

[0059] Dot pattern characteristics: Under fluorescence, reconstructed amber may show many small dot fluorescence. This is due to the impurities in the amber powder or fragments during the pressing process or the organic dyes contained in the colorant, which cause special reactions under fluorescence. There is also a very small probability that it is caused by the distribution of tiny bubbles and voids inside.

[0060] Since there are certain differences in fluorescence characteristics between natural amber and reconstructed amber, the fluorescence characteristics can be used to determine whether the amber to be tested is natural amber.

[0061] In addition, it can be understood that the amber to be tested can be identified by only one of the above-mentioned surface magnification feature, light-transmitting structure feature and fluorescence feature, and the identification efficiency under this embodiment is relatively high. Of course, the amber to be tested can also be identified together by at least two features, and the identification accuracy under this embodiment is relatively high.

[0062] The above scheme can identify reconstructed amber through at least one of the surface magnification characteristics, light-transmitting structure characteristics and fluorescence characteristics. On the one hand, it can make the above-mentioned reconstructed amber identification method applicable to more application scenarios, which is beneficial to improving the adaptability of the above-mentioned reconstructed amber identification method; on the other hand, since there are certain differences between reconstructed amber and natural amber in surface magnification characteristics, light-transmitting structure characteristics and fluorescence characteristics, identifying reconstructed amber through at least one of the surface magnification characteristics, light-transmitting structure characteristics and fluorescence characteristics is beneficial to improving the identification accuracy of the above-mentioned reconstructed amber identification method.

[0063] The following is an introduction to the scheme of obtaining the amber morphological characteristic image of the amber to be tested in step S110:

[0064] The embodiment of the present application can extract the morphological feature image of the amber to be tested by the image feature extraction model, and the image feature extraction model can select pre-trained models such as AlexNet and VGG as the basis for feature extraction to ensure that the universal features of the image can be captured. After completing the pre-training, the pre-trained model can also be fine-tuned to adapt it to the extraction task of the amber morphological feature image, and the stable fitting of the matching value is achieved by adjusting the layer weight. The input of the image feature extraction model can be the amber image collected by the image collector, and the output can be the amber morphological feature image of the amber to be tested and / or the amber sample.

[0065] The above image feature extraction model mainly includes:

[0066] Multi-layer convolutional layers: The hierarchical features of an image can be automatically learned through multi-layer convolutional layers, where low-level convolutional layers are usually used to capture basic edge and texture information, while deep convolutional layers are used to identify more complex patterns and object parts. The convolutional layer uses a convolution kernel to slide on the image and perform weighted summation of local areas to extract image features. This process does not depend on the specific location of the features in the image, so it is translation invariant;

[0067] Activation function layer: Introducing activation functions such as ReLU to increase the nonlinear ability of the model, allowing the model to learn more complex feature representations; activation functions help the network deal with nonlinear problems and enhance the expressiveness of the model;

[0068] Pooling layer: Downsample the feature map through a pooling layer (such as a maximum pooling layer or an average pooling layer) to reduce the feature dimension while retaining the most important information; the pooling operation increases the abstraction level of the feature and enhances the invariance to small deformations;

[0069] Fully connected layer: After the convolutional layer and the pooling layer, the fully connected layer maps the learned distributed features to a high-dimensional space. Regularization techniques such as dropout are usually used in these layers to prevent overfitting.

[0070] The main parameters of the above convolutional layer include:

[0071] Stride: It is used to control the step size of the convolution kernel movement, which affects the size of the output feature map. A larger step size will lead to downsampling and reduce the resolution of the feature map;

[0072] Zero-Padding: Add zero values ​​to the edges of the image to maintain the size of the feature map or allow the convolution kernel to operate at the edges; the padding size is usually 1 so that the output feature map has the same size as the input image;

[0073] Convolution kernel size: Commonly used convolution kernel sizes are 3x3 or 5x5, which can capture features within a local neighborhood while maintaining computational efficiency.

[0074] In addition, it can be understood that the above-mentioned step S110 and the construction process of the atlas database can both use the above-mentioned image feature extraction model to extract feature images.

[0075] The following is an introduction to the scheme of matching the amber morphological characteristic image with the sample characteristic image in the above step S120:

[0076] It is understandable that the image types included in the above-mentioned atlas database exist in the following three implementation modes:

[0077] The first implementation method: the atlas database only contains sample characteristic images of natural amber samples;

[0078] In this implementation mode, the amber morphological characteristic image of the amber to be tested is matched with the sample characteristic image of the natural amber sample in the atlas database. If the degree of matching with the sample characteristic image of the natural amber sample is higher than the first preset matching degree threshold, the amber to be tested is determined to be natural amber; otherwise, the amber to be tested is determined to be reconstructed amber.

[0079] The second implementation method: the atlas database only contains sample characteristic images of the reconstructed amber samples;

[0080] In this implementation mode, the amber morphological characteristic image of the amber to be tested is matched with the sample characteristic image of the reconstructed amber sample in the atlas database. If the degree of matching with the sample characteristic image of the reconstructed amber sample is higher than a second preset matching degree threshold, the amber to be tested is determined to be reconstructed amber; otherwise, the amber to be tested is determined to be natural amber.

[0081] A third implementation method: the atlas database contains sample characteristic images of natural amber samples and reconstructed amber samples;

[0082] In this implementation mode, the above matching scheme has at least the following two implementation modes:

[0083] Implementation method 1: firstly, the amber morphological characteristic image of the amber to be tested is matched with the sample characteristic image of the natural amber sample in the atlas database. If the matching degree with the sample characteristic image of the natural amber sample is higher than the first preset matching threshold, the amber to be tested is determined to be natural amber, and the identification process is terminated;

[0084] If the degree of match with the sample characteristic image of the natural amber sample is not higher than the first preset matching threshold, the amber morphological characteristic image of the amber to be tested is matched with the sample characteristic image of the reconstructed amber sample in the atlas database; if the degree of match is higher than the second preset matching threshold, the amber to be tested is determined to be reconstructed amber;

[0085] If the degree of match between the amber morphological characteristic image of the amber to be tested and the sample characteristic image of the reconstructed amber sample in the atlas database is not higher than the second preset matching threshold, manual intervention can be requested to obtain the identification result through manual judgment.

[0086] Embodiment 2: Match the amber morphological characteristic image of the amber to be tested with the sample characteristic images of natural amber samples and the sample characteristic images of reconstructed amber samples in the atlas database respectively. If the matching degree between the amber morphological characteristic image of the amber to be tested and the sample characteristic image of the natural amber sample in the atlas database is higher than the first preset matching threshold, and the matching degree between the amber morphological characteristic image of the amber to be tested and the sample characteristic image of the reconstructed amber sample in the atlas database is not higher than the second preset matching threshold, then the amber to be tested is determined to be natural amber; if the matching degree between the amber morphological characteristic image of the amber to be tested and the sample characteristic image of the natural amber sample in the atlas database is not higher than the first preset matching threshold, and the matching degree between the amber morphological characteristic image of the amber to be tested and the sample characteristic image of the reconstructed amber sample in the atlas database is higher than the second preset matching threshold, then the amber to be tested is determined to be reconstructed amber; in other cases, the identification result can be obtained by manual judgment.

[0087] In addition, it can be understood that the sample characteristic image in the above-mentioned atlas database can also include at least one of the sample surface magnified characteristic image, the sample light-transmitting structure characteristic image and the sample fluorescence characteristic image of the amber sample. Generally, when performing image matching in the above-mentioned step S120, the characteristic images of the same type of the amber to be tested and the amber sample are matched, for example, the surface magnified characteristic image of the amber to be tested is matched with the surface magnified characteristic image of the amber sample.

[0088] The storage form of the sample characteristic image in the above-mentioned atlas database can be implemented in at least the following two ways:

[0089] The first implementation method: directly storing the sample characteristic image of the amber sample;

[0090] In this implementation manner, the above step S120 can directly implement image matching through an image matching algorithm.

[0091] A second implementation method: storing feature descriptors of sample feature images;

[0092] In this implementation manner, the above-mentioned step S120 includes: obtaining a first feature descriptor of the amber morphological feature image; wherein the first feature descriptor is used to characterize the image features of the amber morphological feature image; matching the first feature descriptor with the second feature descriptor of the sample feature image in the atlas database to obtain the matching degree between the first feature descriptor and the second feature descriptor; wherein the second feature descriptor is used to characterize the image features of the sample feature image;.

[0093] The above feature descriptor is a data structure or vector used to represent image features. In the fields of image processing, computer vision, and deep learning, feature descriptors are usually used to extract and represent key information in images for subsequent image analysis, matching, and recognition tasks. Feature descriptors can be extracted using networks such as Siamese networks and Triplet networks, where:

[0094] Siamese network: A Siamese network is a neural network used to learn a similarity measure for a pair of input images. It consists of two identical network branches that share the same weights. During training, a Siamese network receives a pair of images as input and learns feature representations through a loss function (such as contrastive loss) so that feature vectors of similar images are close to each other in feature space, while feature vectors of dissimilar images are far apart. Siamese networks are often used for feature extraction of a pair of images and can be used for image matching and verification tasks.

[0095] Triplet Network: Triplet Network is another deep learning architecture that learns feature representations by using three image samples (an anchor, a positive sample, and a negative sample). The goal of the Triplet Network is to make the distance between the anchor and the positive sample smaller than the distance between the anchor and the negative sample. The above method is suitable for datasets with clear category labels and can effectively learn feature representations that distinguish different categories.

[0096] It is understandable that various optimization techniques can be used to improve the performance of descriptors during feature descriptor learning. For example, binary descriptors can be used to improve storage efficiency and matching speed. Binary descriptors are implemented by quantizing feature vectors into binary strings, which can be achieved by introducing a binary normalization layer in the network, as adopted in the ZippyPoint method. In addition, in some cases, an end-to-end network architecture can be designed to directly learn feature descriptors of images. This approach usually involves integrating all the steps of feature extraction, feature encoding, and feature matching into a unified network and optimizing the weights of the entire network through backpropagation. When training feature descriptors, different training strategies can be adopted, such as using hard negative sample mining to select samples that are difficult to match correctly for training, or using online sample selection to dynamically update the training sample set.

[0097] In the implementation process of the above scheme, on the one hand, the feature descriptor can capture key information in the image such as texture, shape, color, etc., and the matching between feature descriptors is helpful to improve the identification accuracy of the above-mentioned reconstructed amber identification method; on the other hand, the feature descriptor usually has a lower dimension, which is helpful to reduce the amount of calculation during image matching, thereby helping to improve the identification efficiency of the above-mentioned reconstructed amber identification method.

[0098] As an optional implementation of the above-mentioned reconstructed amber identification method, the above-mentioned step S110 includes: obtaining a multi-scale amber morphological feature image of the amber to be tested.

[0099] At this time, the above-mentioned step S120 includes: obtaining a first feature descriptor of a multi-scale amber morphological feature image; fusing the multi-scale first feature descriptors to obtain a third feature descriptor; matching the third feature descriptor with the second feature descriptor of the sample feature image in the atlas database to obtain the matching degree between the third feature descriptor and the second feature descriptor.

[0100] Exemplarily, the multi-scale amber morphological feature map may be amber morphological feature maps of multiple resolutions. A comprehensive feature descriptor containing multi-scale information is obtained by aggregating the first feature descriptors of the multi-scale amber morphological feature image in the feature space.

[0101] The above scheme obtains a third feature descriptor containing multi-scale information by fusing multi-scale first feature descriptors, and uses the third feature descriptor for image matching, which is beneficial to improving the matching accuracy, thereby improving the identification accuracy of the above-mentioned reconstructed amber identification method.

[0102] The following is an introduction to the scheme for performing image matching in step S120 above:

[0103] In the above step S120, an image matching algorithm such as the FLANN (Fast Library for Approximate Nearest Neighbors) algorithm can be selected to perform the image matching task. The FLANN algorithm can be used to quickly search for the nearest neighbor point in the feature vector set. The FLANN algorithm accelerates the search process by constructing a tree-based data structure, which includes data structures such as kd-tree and hierarchical clustering tree. The accuracy and speed of the search results are controlled by adjusting the parameters of the FLANN algorithm, such as the number of recursive traversals (checks parameter). In addition, a ratio test is used to delete outliers, that is, to compare the distance ratio of the nearest neighbor and the next nearest neighbor to determine the quality of the matching point. Generally, if the ratio of the nearest neighbor distance to the next nearest neighbor distance is less than a set threshold (such as 0.7), the match is considered to be valid.

[0104] It is understandable that the above image matching process can also be combined with the RANSAC algorithm for robust matching to filter out incorrect matching points. The RANSAC algorithm estimates model parameters by iteratively randomly selecting point pairs and evaluating the consistency of all point pairs to identify inliers and outliers.

[0105] In addition, during the image matching process, the homography matrix can also be used to geometrically transform the matching points to correct the geometric distortion between images. This usually involves calculating the perspective transformation of the matching point pairs and then applying the transformation to align the images.

[0106] Exemplarily, the above-mentioned image matching process can also improve the image matching speed and the ability to process large-scale data sets by selecting appropriate data structures and indexing strategies and adopting parallel computing technology.

[0107] The following is an introduction to the scheme for obtaining the spectrum identification result in step S130:

[0108] Exemplarily, the above-mentioned atlas identification result may include: an identification result of whether the amber to be tested is natural amber. For the method of obtaining the identification result, please refer to the introduction of the atlas database construction method in the above content.

[0109] The above-mentioned spectrum identification result may also include: when the amber to be tested is reconstructed amber, the type of the amber to be tested is:

[0110] As an optional implementation of the above-mentioned reconstructed amber identification method, the above-mentioned reconstructed amber identification method further includes: when the spectrum identification result is reconstructed amber, determining the reconstructed amber type of the amber to be tested according to the matching degree. This implementation example: after determining that the amber to be tested is reconstructed amber, the type of the amber to be tested can be determined by the matching degree between the amber morphological characteristic image of the amber to be tested and the sample characteristic images of different reconstructed amber samples, for example, the reconstructed amber type with the highest matching degree is determined as the type of the amber to be tested.

[0111] The above scheme can not only determine whether the amber to be tested is natural amber, but also determine the type of amber to be tested when it is reconstructed amber, so that the above reconstructed amber identification method can be applied to more application scenarios, which is conducive to improving the adaptability of the above reconstructed amber identification method.

[0112] As an optional implementation of the above-mentioned reconstructed amber identification method, after step S110, the above-mentioned reconstructed amber identification method further includes: performing at least one of denoising, contrast enhancement, size standardization and edge enhancement on the amber morphological feature image. This implementation, for example, performs non-mean filtering or sharpening on the surface magnified feature image to improve the details and sharpness of the image, performs histogram equalization on the light-transmitting structure feature image to enhance the light-dark contrast of the image, and marks and colors the fluorescence feature image to better observe the specific fluorescence features.

[0113] The above scheme improves the matching accuracy of image matching by performing at least one of denoising, contrast enhancement, size standardization and edge enhancement on the amber morphological feature image, thereby improving the identification accuracy of the above-mentioned reconstructed amber identification method.

[0114] Based on the same inventive concept, an image acquisition method is also provided in an embodiment of the present application, which method includes: extracting sample morphological features from the morphological image of the amber sample to obtain a sample morphological feature image; wherein the sample morphological features include: at least one of sample surface magnification features, sample light-transmitting structure features and sample fluorescence features; the sample morphological feature image is used to construct a map database.

[0115] As an optional implementation of the above-mentioned image acquisition method, the above-mentioned image acquisition method also includes: obtaining a sample morphological feature descriptor of the sample morphological feature image; wherein the sample morphological feature descriptor is used to construct an atlas database.

[0116] As an optional implementation of the above-mentioned image acquisition method, extracting sample morphological features from the morphological image of the amber sample includes: extracting sample morphological features from the morphological image of the natural amber sample and / or the reconstructed amber sample.

[0117] As an optional implementation manner of the above-mentioned image acquisition method, the sample morphological features are extracted in the morphological image of the reconstructed amber sample, including: extracting the sample morphological features in the morphological image of at least one of the reconstructed amber samples selected from the group consisting of reconstructed Russian white flower honey, reconstructed full wax white flower golden twisted honey, reconstructed golden twisted honey, reconstructed half mountain and half water flower honey, reconstructed chicken oil yellow beeswax, reconstructed roasted old beeswax, imitation blood beeswax, imitation Burmese flower amber, and imitation Fushun root amber.

[0118] The following is a solution for building the above graph database in a certain application scenario, which mainly includes:

[0119] Step 1: Obtain amber samples;

[0120] Amber samples may include natural amber and / or reconstructed amber, wherein the acquisition scheme of reconstructed amber mainly includes:

[0121] (1) Obtaining raw materials: Collecting amber fragments or powder, using scraps or small pieces of amber from the processing of amber handicrafts as the main materials. Sometimes, semi-petrochemical resins (copal resins) are added to reduce costs. For example, dyes (chromium oxide, organic dyes) and curing agents (epoxy resins, phenolic resins) are added to make artificial amber. For example, beeswax is added to make beeswax.

[0122] (2) Melting resin: amber fragments or powder mixed with copal resin in a ratio of 2:3 are added to a melting pot, and heated in stages for 40-120 minutes, starting at 70°C and gradually heated to 150°C to fully melt it. When the molten resin can fully flow, the heating temperature is fixed, and the large bubbles appearing due to boiling gradually turn into small bubbles with a diameter of less than 0.5 mm, and then the heating is continued for 15 minutes. The staged heating is stopped, and the resin is placed in a hot air circulation furnace and kept at 100°C to maintain a semi-melted state;

[0123] (3) Prepare coloring materials: Due to the different characteristics of the origin of molten resin fragments and copal resin, coloring materials need to be prepared according to the state of the two molten resins. If the staged heating time is long, multiple units of epoxy resin are mixed to stabilize the curing effect. If the melting temperature is high, multiple units of phenolic resin are mixed to bond the internal structure. If special colors such as white honey or chicken oil yellow are prepared, corresponding dyes need to be added. If beeswax is prepared, beeswax is added and the content ratio is prepared according to the target effect. The prepared coloring materials are mixed and stirred with the corresponding curing agent as the base, and heated to 60°C. The whole process needs to be protected from light;

[0124] (4) Mixing raw materials: Mixing raw materials determines the uniformity and quality of the final product and is also the link in producing the internal structural characteristics of reconstructed amber. First, prepare the semi-melted molten resin at 100°C and the colorant at 60°C. Next, slowly pour the molten molten resin into a preheated container (the temperature must be higher than the melting point of the colorant), and then add the colorant slowly to avoid splashing or generating large bubbles. Use a stirrer for preliminary stirring, then replace with a smaller stirring rod and continue stirring until the two materials are completely mixed to form a uniform mixture. During this process, it is necessary to constantly monitor the temperature of the mixture to ensure that it remains within the appropriate range, and to heat appropriately when necessary to maintain fluidity. At the same time, pay attention to eliminating large bubbles that may be generated during the mixing process, which can be achieved by tapping the outer wall of the container or using a vacuum device. Finally, check the quality of the mixture to ensure that it is free of particles, separation, and shows good uniformity and consistency. When the mixture reaches the ideal state, the next step of pouring and molding can be carried out.

[0125] (5) Pressing: First, slowly pour the mixed raw material mixture into the prepared mold to ensure that the mixture can cover the entire surface of the mold and fill every detail. Currently, a round mold is generally used, and cutting is performed after molding. Subsequently, the mold containing the mixture is placed under the press, and pressure is applied slowly and evenly. At the same time, the mold is heated by an external heating source so that the temperature inside the mold gradually rises to the processing temperature of 120°C. During the heating process, the softening and flow of the material need to be closely observed to ensure that the mixture is evenly distributed in the mold and no bubbles or voids are generated. As the material gradually softens, the pressure is gradually increased until the final temperature of 135°C and the final pressure of 1200pa are reached, which are predetermined according to the size and shape of the pressed part.

[0126] (6) Demolding and Cooling: After the pressing process is completed, when the pressure in the mold is released and the temperature has not dropped significantly, this is the best time to demold. At this point, the resin and beeswax mixture of the pressed part has solidified into shape, but still retains a certain degree of thermoplasticity, allowing the pressed part to be smoothly removed from the mold without damage. The demolding process requires careful operation to ensure that every detail of the pressed part can be completely separated from the mold without leaving any burrs or cracks. During the entire cooling process, the pressed part will gradually solidify and eventually form a hard and stable reconstructed amber product.

[0127] See also Figure 2 The following introduces the ingredients and processing methods of nine types of reconstructed amber: reconstructed Russian white flower honey, reconstructed full wax white flower golden twisted honey, reconstructed golden twisted honey, reconstructed half mountain and half water flower honey, reconstructed chicken oil yellow beeswax, reconstructed roasted old beeswax, imitation blood beeswax, imitation Burmese flower amber, and imitation Fushun root amber:

[0128] (1) Reconstructed Russian white flower honey (see Figure 2 Chinese number: Zz01~Zz05);

[0129] Ingredients: 3 units of amber chunks, 2 units of copal resin, 1 unit of epoxy resin, 1 unit of phenolic resin, 0.2 units of chromium oxide, 0.1 units of organic dye (lemon yellow), 12.7 units of beeswax.

[0130] Processing method: melt at 150°C for 70 minutes, then mix and heat at 230°C, 1600pa pressure, and vacuum cool and solidify.

[0131] (2) Re-making wax white flower golden honey (see Figure 2 Chinese number: Zz06~Zz10);

[0132] Ingredients: 2 units of amber small pieces, 1 unit of copal resin, 2 units of epoxy resin, 0.5 units of organic dye (gold), 15 units of beeswax.

[0133] Processing method: melt at 160℃ for 80 minutes, then mix and heat at 240℃, 1800pa pressure, and vacuum cool and solidify.

[0134] (3) Reconstructed golden honey (see Figure 2 Chinese number: Zz11~Zz15);

[0135] Ingredients: 4 units of amber fragments, 1 unit of copal resin, 0.5 units of epoxy resin, 0.3 units of organic dye (gold), 10 units of beeswax.

[0136] Processing method: melt at 155°C for 75 minutes, then mix and heat at 235°C, 1500pa pressure, and vacuum cool and solidify.

[0137] (4) Recreate half mountain and half water nectar (see Figure 2 Chinese number: Zz16~Zz20);

[0138] Ingredients: 2 units of amber scraps, 2 units of copal resin, 1 unit of epoxy resin, 0.2 units of organic dye (gold), 10 units of beeswax.

[0139] Processing method: melt at 140℃ for 60 minutes, then mix and heat at 180℃ and 210℃ in two stages according to the size of bubbles generated, slowly pressurize to 1400pa multiple times during vacuuming, keep warm during cooling and solidification, and maintain vacuum cooling to 40℃ at an average speed of 12 hours.

[0140] (5) Reconstructed chicken fat yellow beeswax (see Figure 2 Chinese number: Zz21~Zz25);

[0141] Ingredients: 3 units of amber crushed materials, 1 unit of copal resin, 1 unit of epoxy resin, 2 units of phenolic resin, 0.2 units of organic dye (chicken fat yellow), and 12 units of beeswax.

[0142] Processing method: Melt at 130℃ for 120 minutes, quickly heat to 180℃ after the scraps are completely melted, mix the coloring material, then mix and heat at 260℃, and solidify by vacuum cooling at a pressure of 1700pa.

[0143] (6) Re-creation of baked old beeswax (see Figure 2 Chinese number: Zz26~Zz30);

[0144] Ingredients: 3 units of amber crushed materials, 3 units of copal resin, 1 unit of epoxy resin, 0.1 unit of organic dye (sunset yellow), 0.2g thickener, 12 units of beeswax.

[0145] Processing method: melt at 150°C for 70 minutes, add coloring material, mix and heat at 330°C, 1500pa pressure, and vacuum cool and solidify.

[0146] (7) Imitation blood beeswax (see Figure 2 Chinese number: Zz31~Zz35);

[0147] Ingredients: 2 units of amber scraps, 2 units of copal resin, 1 unit of epoxy resin, 0.3 units of sunset red dye, 10 units of beeswax.

[0148] Processing method: melt at 150°C for 70 minutes, add colorants, mix and heat at 300°C, 1600pa pressure, heat and cool multiple times, add raw materials and colorants in batches, and cool and solidify in vacuum.

[0149] (8) Imitation Burmese amber (see Figure 2 Medium number: Zz36~Zz40);

[0150] Ingredients: 4 units of amber crushed materials, 0.5 units of copal resin, 1.5 units of epoxy resin, 0.1 units of organic dye (honey color), 8 units of beeswax.

[0151] Processing method: Melt at 150℃ for 20 minutes and quickly heat to 220℃, the color will increase rapidly, add coloring materials, mix and heat at 220℃, 1400pa pressure, and vacuum cool and solidify.

[0152] (9) Imitation Fushun amber (number: Zz41~Zz45);

[0153] Ingredients: 2 units of amber crushed materials, 0.5 units of copal resin, 5 units of epoxy resin, 2 units of phenolic resin, 0.1 units of organic dye (honey color), 8 units of beeswax.

[0154] Processing method: Melt at 150℃ for 20 minutes and quickly heat to 220℃, the color will increase rapidly, add coloring materials, mix and heat at 220℃, 1400pa pressure, and vacuum cool and solidify.

[0155] Step 2, obtaining a sample characteristic image of the amber sample;

[0156] (1) obtaining a sample surface enlarged characteristic image of the amber sample;

[0157] Place the amber samples Zz01 to Zz45 to be tested on a stable and dust-free operating table to ensure that the sample surface is clean. Use a ring light to illuminate the amber sample at a 90-degree vertical angle. Set the image collector at 45 degrees above the top. Adjust the image collector to a 30x magnification angle. Set the f1.8 aperture, 1 / 100 second shutter, ISO 100, fixed focus, turn off the dynamic adjustment gear, automatically synthesize a high depth of field image, transfer the image to the host computer, obtain the amber surface magnified image, and then use the feature extraction model to extract the sample surface magnified feature image;

[0158] (2) obtaining a sample light transmission structure characteristic image of the amber sample;

[0159] Place the light source vertically close to the outer surface of the amber sample to be tested, ensure that the light penetrates the sample vertically at 90 degrees, and face the image collector directly to the translucent amber sample, maintain a 30x magnification angle, f1.8 aperture, 1 / 1200 second shutter, ISO 50, fixed focus, turn off dynamic adjustment gear, and automatically synthesize high depth of field images. Transmit the image to the host computer to obtain the amber translucent structure image, and then use the feature extraction model to extract the amber translucent structure feature image.

[0160] (3) obtaining a sample fluorescence characteristic image of the amber sample;

[0161] In a dark field environment, a 365-nanometer ultraviolet light source was used to replace the above-mentioned illumination source, perpendicular to the sample surface and maintained at a height of more than 10 cm. The image collector maintained a 45-degree angle of refracted light to observe the outer surface of the amber. The magnification angle was set to 30 times, the aperture was f1.8, the shutter was 1 / 150 seconds, the ISO was 150, the focus was fixed, the dynamic adjustment gear was turned off, the high depth of field image was automatically synthesized, the image was transmitted to the host computer, the amber fluorescence image was obtained, and then the feature extraction model was used to extract the amber fluorescence feature image.

[0162] Step 3: Image preprocessing;

[0163] The sample surface magnified feature image and the fluorescence feature image obtained in step 2 are subjected to contrast enhancement, edge enhancement, multi-image synthesis, non-mean filter sharpening and other processing to improve image details, obtain a processed grayscale image, and use a dark-marked fluorescence image to mix the fluorescence magnification feature and the surface texture feature. The sample surface magnified feature image, the sample light-transmitting structure feature image and the sample fluorescence feature image can be fused into a sample feature image. The synthesized image is, for example, Figure 3 Img-Zz01~Img-Zz45.

[0164] Step 4: Database entry;

[0165] Import the processed sample feature images into the atlas database, annotate the images, and record the sample information and acquisition conditions.

[0166] Depend on Figure 3 The sample characteristics can be seen in:

[0167] (1) The surface of the reconstructed Russian white flower honey (No.: Zz01~Zz05) has abnormal flowing texture, and the angle and stacking characteristics have a reconstructed extrusion state. Because the air pressure is not high, there is no distortion, the overall transition is relatively smooth, and there is no obvious sharp angle. Under the fluorescent state, due to the use of epoxy resin and beeswax for coloring and curing, blood-streaked fluorescence characteristics appear. Under magnified observation, because it uses large pieces of amber fragments for melting, obvious broken spots can be seen.

[0168] (2) The surface of the reconstructed full wax white flower golden twisted honey (No.: Zz06~Zz10) has some flow textures caused by processing. Due to the small amber pieces in the ingredients and the high content of beeswax, the texture is more delicate than other reconstructed beeswax. The overall transition is relatively smooth, and there are no obvious sharp angles. Under the fluorescent state, due to the use of epoxy resin and gold organic dyes, it shows fluorescence characteristics related to gold. Under magnified observation, it can be found that due to the use of small pieces of amber pieces for melting, there are small fragments of combined traces.

[0169] (3) The surface texture of the reconstructed gold twisted honey (number: Zz11~Zz15) shows a dot-like reconstruction feature. Due to the high content of amber fragments, the texture may be more obvious and have a certain directionality. The overall transition is relatively natural, but traces of reconstruction and extrusion can still be seen. In the fluorescent state, due to the presence of organic dyes, corresponding yellow-green spot-like fluorescence features will appear. By magnifying its internal structure and observing it, traces left by the fusion of fragments during the processing process can be seen.

[0170] (4) The surface texture of the reconstructed half-mountain and half-water nectar (No.: Zz16-Zz20) presents a unique extrusion blood streak due to the staged heating and multiple vacuuming and pressurization during the processing process. There are different degrees of bubble marks reflected in the texture. The overall transition has a certain sense of hierarchy, and there are no obvious sharp angles. Under the fluorescent state, due to the presence of organic dyes, there are corresponding patchy fluorescence performances. Through image processing to analyze its internal structure, traces caused by the heating and pressurization process and signs of fusion of broken materials can be seen.

[0171] (5) The surface of the reconstructed chicken-fat yellow beeswax (No.: Zz21~Zz25) has a filamentous strip texture due to processing. Due to the influence of multiple temperature stages and ingredient ratios during the processing, the texture is relatively complex, and the details are diverse and have different identification characteristics, but the overall transition is relatively smooth. In some parts, there are slight changes due to rapid heating and ingredient fusion. In the fluorescent state, due to the use of organic dyes, it presents a specific patchy fluorescent feature. By observing the internal structure, traces left by the melting of crushed materials and the mixing of ingredients can be found.

[0172] (6) The surface texture of reconstructed old beeswax (No.: Zz26~Zz30) shows a convex expansion and extension filling feature due to high temperature baking and processing. It may have a similar skin color texture to old beeswax, but careful observation can reveal traces of reconstructed fractures and abnormal texture angles. The overall transition is relatively natural, but it is different from natural old beeswax in color transition and texture details. In the fluorescent state, due to the presence of sunset yellow dye and the influence of the processing process, it has a unique fragmented fluorescent performance.

[0173] (7) The surface texture of imitation blood beeswax (No.: Zz31~Zz35) shows a unidirectional strip arrangement different from that of natural blood beeswax due to the processing process. It is more regular and lacks the natural texture extension and flow of natural blood beeswax. The overall transition is too smooth and the color transition is too abrupt. In the fluorescent state, due to the presence of sunset red dye, it shows a fragmented fluorescence characteristic different from that of natural blood beeswax. By observing the internal structure, traces left by the mixing of raw materials and the processing process can be found.

[0174] (8) The surface texture of imitation Burmese amber (No.: Zz36~Zz40) presents a unique egg-shaped feature due to the processing process, which is obviously different from natural Burmese amber. The overall transition is unnatural, and there are obvious differences in texture details and color from natural products. In the fluorescent state, due to the presence of honey-colored dye, there is a corresponding strong fluorescence performance. By observing the internal structure, traces left by the mixing of raw materials and the processing process can be found.

[0175] (9) The surface texture of the imitation Fushun root amber (No.: Zz40~Zz45) shows an egg-shaped feature different from the natural Fushun root amber due to the processing process. It is more regular and lacks the natural texture changes of the natural Fushun root amber. The overall transition is unnatural, but it is different from the natural product in color transition and the transformation of the original amber fragments. In the fluorescent state, it has a corresponding strong fluorescence due to the presence of honey-colored dye. By observing the internal structure, traces left by the mixing of raw materials and the processing process can be found.

[0176] See also Figure 4 Based on the same inventive concept, the present application embodiment provides a reconstructed amber identification device 200, which includes:

[0177] An amber morphological characteristic image acquisition module 210 is used to acquire an amber morphological characteristic image of the amber to be tested;

[0178] A matching degree acquisition module 220 is used to match the amber morphological characteristic image with the sample characteristic image in the atlas database to obtain the matching degree between the amber morphological characteristic image and the sample characteristic image;

[0179] The spectrum identification result acquisition module 230 is used to obtain the spectrum identification result of the amber to be tested according to the matching degree; wherein the spectrum identification result is used to indicate whether the amber to be tested is natural amber.

[0180] As an optional implementation of the above-mentioned reconstructed amber identification device, the above-mentioned amber morphological characteristic image acquisition module 210 is specifically used to: obtain at least one of the surface magnified characteristic image, the transparent structure characteristic image and the fluorescence characteristic image of the amber to be tested.

[0181] As an optional implementation of the above-mentioned reconstructed amber identification device, the above-mentioned matching degree acquisition module 220 is specifically used to: obtain a first feature descriptor of the amber morphological feature image; match the first feature descriptor with the second feature descriptor of the sample feature image in the atlas database, and obtain the matching degree between the first feature descriptor and the second feature descriptor.

[0182] As an optional implementation of the above-mentioned reconstructed amber identification device, the above-mentioned amber morphological characteristic image acquisition module 210 is specifically used to: acquire a multi-scale amber morphological characteristic image of the amber to be tested;

[0183] At this time, the above-mentioned matching degree acquisition module 220 is specifically used to: obtain the first feature descriptor of the multi-scale amber morphological feature image; fuse the first feature descriptors of the multi-scale to obtain the third feature descriptor; match the third feature descriptor with the second feature descriptor of the sample feature image in the atlas database to obtain the matching degree between the third feature descriptor and the second feature descriptor.

[0184] As an optional implementation of the above-mentioned reconstructed amber identification device, the above-mentioned reconstructed amber identification device also includes:

[0185] The reconstructed amber type determination module is used to determine the reconstructed amber type of the amber to be tested according to the matching degree when the spectrum identification result is reconstructed amber.

[0186] As an optional implementation of the above-mentioned reconstructed amber identification device, the above-mentioned reconstructed amber identification device also includes:

[0187] The image preprocessing module is used to perform at least one of denoising, contrast enhancement, size standardization and edge enhancement on the amber morphological feature image.

[0188] Figure 5 A schematic diagram of an electronic device provided in an embodiment of the present application. Figure 5 The electronic device 300 includes: a processor 310, a memory 320 and a communication interface 330. These components are interconnected and communicate with each other through a communication bus 340 and / or other forms of connection mechanisms (not shown).

[0189] The memory 320 includes one or more (only one is shown in the figure), which may be, but not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc. The processor 310 and other possible components may access the memory 320 and read and / or write data therein.

[0190] The processor 310 includes one or more (only one is shown in the figure), which can be an integrated circuit chip with signal processing capabilities. The above-mentioned processor 310 can be a general-purpose processor, including a central processing unit (CPU), a micro controller unit (MCU), a network processor (NP) or other conventional processors; it can also be a special-purpose processor, including a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0191] The communication interface 330 includes one or more (only one is shown in the figure), which can be used to communicate directly or indirectly with other devices to exchange data. For example, the communication interface 330 can be an Ethernet interface; it can be a mobile communication network interface, such as an interface of a 3G, 4G, or 5G network; or it can be other types of interfaces with data transceiving functions.

[0192] One or more computer program instructions may be stored in the memory 320, and the processor 310 may read and execute these computer program instructions to implement the reconstructed amber identification method or image acquisition method and other desired functions provided in the embodiments of the present application.

[0193] Understandably, Figure 5 The structure shown is for illustration only. The electronic device 300 may also include Figure 5 More or fewer components as shown, or with Figure 5 Different configurations are shown. Figure 5 Each component shown in the figure can be implemented by hardware, software or a combination thereof. For example, the electronic device 300 can be a single server (or other device with computing processing capabilities), a combination of multiple servers, a cluster of a large number of servers, etc., and can be both a physical device and a virtual device.

[0194] It is understandable that the above-mentioned reconstructed amber identification method and image acquisition method can be deployed in the same electronic device 300, or in different electronic devices 300.

[0195] The present application also provides a computer-readable storage medium, which stores computer program instructions. When the computer program instructions are read and executed by a computer processor, the reconstructed amber identification method or image acquisition method provided by the present application is executed. For example, the computer-readable storage medium can be implemented as Figure 5 The memory 320 in the electronic device 300.

[0196] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0197] In addition, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0198] Furthermore, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0199] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for identifying reconstructed amber, characterized in that: The method comprises: Acquire an amber morphological characteristic image of the amber to be tested; Matching the amber morphological characteristic image with the sample characteristic image in the atlas database to obtain the matching degree between the amber morphological characteristic image and the sample characteristic image; According to the matching degree, a spectrum identification result of the amber to be tested is obtained; wherein the spectrum identification result is used to characterize whether the amber to be tested is natural amber.

2. The method for identifying reconstructed amber according to claim 1, wherein: The step of obtaining the amber morphological characteristic image of the amber to be tested comprises: Obtain at least one of the surface magnified characteristic image, the light-transmitting structure characteristic image and the fluorescence characteristic image of the amber to be tested.

3. The method for identifying reconstructed amber according to claim 1, wherein: The step of matching the amber morphological characteristic image with a sample characteristic image in a graph database to obtain a matching degree between the amber morphological characteristic image and the sample characteristic image includes: Acquire a first feature descriptor of the amber morphological feature image; wherein the first feature descriptor is used to characterize image features of the amber morphological feature image; The first feature descriptor is matched with the second feature descriptor of the sample feature image in the atlas database to obtain the matching degree between the first feature descriptor and the second feature descriptor; wherein the second feature descriptor is used to characterize the image features of the sample feature image.

4. The method for identifying reconstructed amber according to claim 1, wherein: The step of obtaining the amber morphological characteristic image of the amber to be tested comprises: Acquire multi-scale amber morphological characteristic images of the amber to be tested; The step of matching the amber morphological characteristic image with a sample characteristic image in a graph database to obtain a matching degree between the amber morphological characteristic image and the sample characteristic image includes: Acquire a first feature descriptor of the multi-scale amber morphological feature image; Fusing the first feature descriptors at multiple scales to obtain a third feature descriptor; The third feature descriptor is matched with the second feature descriptor of the sample feature image in the atlas database to obtain the matching degree between the third feature descriptor and the second feature descriptor.

5. The method for identifying reconstructed amber according to any one of claims 1 to 4, characterized in that: The method further comprises: When the atlas identification result is reconstructed amber, the type of reconstructed amber of the amber to be tested is determined according to the matching degree.

6. The method for identifying reconstructed amber according to any one of claims 1 to 4, characterized in that: After obtaining the amber morphological characteristic image of the amber to be tested, the method further comprises: The amber morphological feature image is subjected to at least one of denoising, contrast enhancement, size standardization and edge enhancement.

7. An image acquisition method, characterized in that: The method comprises: The sample morphological features are extracted from the morphological image of the amber sample to obtain the sample morphological feature image; wherein the sample morphological features include: at least one of the sample surface magnification feature, the sample light-transmitting structure feature and the sample fluorescence feature; the sample morphological feature image is used to construct an atlas database.

8. The image acquisition method according to claim 7, characterized in that: The method further comprises: A sample morphological feature descriptor of the sample morphological feature image is obtained; wherein the sample morphological feature descriptor is used to construct the atlas database.

9. The image acquisition method according to claim 7 or 8, characterized in that: The step of extracting sample morphological features from the morphological image of the amber sample comprises: Extract sample morphological features from morphological images of natural amber samples and / or reconstructed amber samples.

10. The image acquisition method according to claim 9, characterized in that: Extracting sample morphological features from the morphological image of the reconstructed amber sample includes: The sample morphological features are extracted from the morphological images of at least one reconstructed amber sample including reconstructed Russian white flower honey, reconstructed full wax white flower golden twisted honey, reconstructed golden twisted honey, reconstructed half mountain and half water flower honey, reconstructed chicken oil yellow beeswax, reconstructed roasted old beeswax, imitation blood beeswax, imitation Burmese flower amber, and imitation Fushun root amber.

11. A reconstructed amber identification device, characterized in that: The device comprises: An amber morphological characteristic image acquisition module is used to acquire an amber morphological characteristic image of the amber to be tested; A matching degree acquisition module, used for matching the amber morphological characteristic image with the sample characteristic image in the atlas database to obtain the matching degree between the amber morphological characteristic image and the sample characteristic image; The spectrum identification result acquisition module is used to obtain the spectrum identification result of the amber to be tested according to the matching degree; wherein the spectrum identification result is used to indicate whether the amber to be tested is natural amber.

12. An electronic device, characterized in that: include: A processor, a memory and a communication bus, wherein the processor and the memory communicate with each other via the communication bus; The memory stores program instructions executable by the processor, and the processor can execute the method according to any one of claims 1 to 6 or 7 to 10 by calling the program instructions.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a computer, the computer is caused to perform the method according to any one of claims 1 to 6 or 7 to 10.

Images