Method and system for identifying authenticity of ancient metal cast coins, terminal and storage medium
By identifying the rusted areas of ancient metal coins, collecting SEM morphology and EDS elemental composition and distribution, and generating rust layer characteristics, the problem of insufficient accuracy in counterfeit identification in existing technologies is solved, and high-accuracy identification of ancient metal coins is achieved.
Patent Information
- Application Number
- CN202511421731.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-30
- Publication Date
- 2025-11-18
AI Technical Summary
Existing technologies struggle to identify counterfeit materials that lack fluorescence when authenticating ancient metal coins, resulting in insufficient accuracy in authentication.
By identifying the rusted areas on the metal coins to be tested, collecting scanning electron microscopy (SEM) morphology and energy dispersive spectroscopy (EDS) elemental composition and distribution, the elemental characteristics, locational characteristics, and crystal growth morphology characteristics of the rust layer are generated. Combining these characteristics, the probability of counterfeit identification is generated, and the SEM morphology and EDS elemental composition and distribution are used to distinguish between artificially induced rust and natural corrosion rust.
It improves the accuracy of identifying counterfeit ancient metal coins, with a genuine coin identification rate of over 93% and a counterfeit coin detection rate of over 88%, which is significantly better than the traditional ultraviolet fluorescence method.
Smart Images
Figure CN120971476A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of cultural relic authentication technology, and in particular to a method, system, terminal and storage medium for identifying counterfeit ancient metal coins. Background Technology
[0002] Ancient coinage is an important cultural relic, playing a vital role in historical and archaeological research, museum collection management, auction appraisal, and judicial evidence collection.
[0003] The relevant technology for authenticating ancient metal coins relies on ultraviolet fluorescence detection. This involves illuminating the coin's surface with an ultraviolet lamp and observing the fluorescence. If no fluorescence reaction occurs, the coin is considered genuine; if it exhibits strong fluorescence or mottled, uneven fluorescence, it is highly likely to be a counterfeit.
[0004] The aforementioned technologies can only identify counterfeit materials made from organic binder-mixed pigments that contain fluorescent components (such as conjugated chemical bonds or aromatic rings). For counterfeit materials that do not have a fluorescent effect, ultraviolet fluorescence detection is ineffective. Summary of the Invention
[0005] To improve the accuracy of identifying counterfeit ancient metal coins, this application provides a method, system, terminal, and storage medium for identifying counterfeit ancient metal coins.
[0006] Firstly, this application provides a method for authenticating ancient metal coins, employing the following technical solution: A method for authenticating ancient metal coins includes: Identify the rusted areas on the metal coin to be tested; The scanning electron microscope (SEM) morphology of the corroded area was collected; The elemental composition and distribution of the corroded area were determined by energy dispersive spectroscopy (EDS). Based on the elemental composition and distribution of EDS, the elemental characteristics of the rust layer of the metal coin under test were obtained; The SEM morphology is segmented and classified to obtain the locational characteristics and crystal growth morphology of the rust layer of the metal coin under test. Based on the elemental characteristics of the rust layer, the locational characteristics of the rust layer, and the morphological characteristics of the rust layer crystal growth, the probability of identifying counterfeit metal coins to be tested is generated.
[0007] By adopting the above technical solution, after obtaining the SEM morphology and EDS elemental composition and distribution of the rusted area of the metal coin to be tested, the elemental characteristics, locational characteristics, and crystal growth morphology characteristics of the rust layer in the rusted area are obtained using the SEM morphology and EDS elemental composition and distribution. Based on these characteristics, the probability of identifying counterfeit metal coins is generated. This technical solution, combined with the characteristics of the rusted area, can quickly distinguish between artificially induced rust and natural corrosion rust, thereby improving the accuracy of identifying counterfeit ancient metal coins.
[0008] Optional, set the detection element; Based on the composition and distribution of the EDS elements, determine whether the detected elements are present in the rusted area; If not, the default value will be used to generate the rust layer element features; If so, then determine the positional distribution of the detected element in the corroded area; Based on the location distribution, the enrichment degree of the detected elements is calculated to obtain the elemental characteristics of the rust layer.
[0009] By adopting the above technical solution, the elemental characteristics of the rust layer are obtained by using the elemental composition and distribution of EDS. These elemental characteristics can determine the enrichment degree of the detected elements and generate rust layer elemental characteristics, so that the rust layer elemental characteristics can more accurately represent the situation of the detected elements in the rust area and improve the accuracy of identifying counterfeits of ancient metal coins.
[0010] Optionally, cluster analysis is performed on the detected elements based on the location distribution to obtain the enriched regions corresponding to the detected elements; Obtain the total number of particles of the detected element in each enrichment region; Calculate the average total number of particles to obtain the average content of the detected elements; Feature extraction is performed on the average content of the detected elements to obtain the elemental characteristics of the rust layer.
[0011] By employing the above technical solution, cluster analysis is performed on the detected elements to obtain enriched regions. The average content of the detected elements is then obtained from these enriched regions, and feature extraction is performed on the average content of the detected elements to obtain the rust layer element characteristics. This allows the rust layer element characteristics to more accurately represent the situation of the detected elements within the rusted area, improving the accuracy of counterfeit identification of ancient metal coins.
[0012] Optionally, based on the SEM morphology, the SEM morphology is segmented to obtain rust layer stacks and rust layer location features, wherein the rust layer location features are used to describe the stacking order of rust layer stacks arranged starting from the surface of the metal coin under test. Based on the SEM morphology, the morphological features of the rust layer stacks are determined, and the morphological features of the rust layer stacks are used to describe the morphology of a single rust layer stack. By integrating the morphological features of the rust layer stacks, the morphological features of the rust layer crystal growth are obtained.
[0013] By employing the above technical solution, the SEM morphology is segmented to obtain the rust layer stacking and rust layer location characteristics. Based on the SEM morphology, the rust layer stacking morphology characteristics are determined. Integrating the rust layer stacking morphology characteristics, the rust layer crystal growth morphology characteristics are obtained. This technical solution makes the rust layer location characteristics and rust layer crystal growth morphology characteristics more consistent with actual conditions, improving the accuracy of counterfeit identification of ancient metal coins.
[0014] Optionally, a first counterfeit detection probability can be obtained based on the elemental characteristics of the rust layer; The similarity between the locational characteristics of the rust layer and the locational characteristics of the standard rust layer is calculated to obtain the second probability of fake detection; In the morphological feature library, the feature difference between the morphological feature of the rust layer crystal growth and other morphological features of the rust layer crystal growth is calculated to obtain the morphological difference feature; The third spoofing probability is obtained based on the minimum value of the difference in the morphological features. The first false detection probability, the second false detection probability, and the third false detection probability are weighted and calculated to obtain the false detection probability.
[0015] By employing the above technical solution, based on the elemental characteristics, locational characteristics, and crystal growth morphology of the rust layer, a first, second, and third false detection probability are generated sequentially. These probabilities are then weighted and calculated to obtain the final false detection probability. This technical solution makes the false detection probability more accurate.
[0016] Optionally, the morphological feature library includes a genuine coin morphological feature library and a counterfeit coin morphological feature library; Take the morphological features of genuine coins from the aforementioned genuine coin morphological feature library; The difference between the morphological characteristics of the rust layer crystal growth and the morphological characteristics of the genuine coin is calculated to obtain the morphological characteristic difference of the genuine coin. Take the counterfeit currency morphology features from the counterfeit currency morphology feature library; The difference between the morphological characteristics of the rust layer crystal growth and the morphological characteristics of the counterfeit coin is calculated to obtain the morphological characteristic difference of the counterfeit coin. The feature difference is obtained by weighted calculation of the difference in appearance features between the genuine coin and the counterfeit coin.
[0017] By adopting the above technical solution, the feature difference values for genuine and counterfeit coins are obtained based on the genuine coin feature library and the counterfeit coin feature library, respectively. A weighted calculation of these two feature difference values generates a more accurate feature difference.
[0018] Optionally, if the probability of detecting a fake is less than a preset probability threshold, a first type of processing suggestion is generated; When the probability of detecting counterfeits is greater than the preset probability threshold, the type of rust in the rusted area is obtained based on the elemental characteristics of the rust layer and the crystal growth morphology characteristics of the rust layer. A second type of treatment recommendation is generated based on the type of corrosion described.
[0019] By adopting the above technical solution, corresponding processing suggestions are generated based on the counterfeit detection probability. Technicians can then perform corresponding operations on the metal coin to be tested in order to protect the metal coin.
[0020] Secondly, this application provides a system for identifying counterfeit ancient metal coins, employing the following technical solution: A system for authenticating ancient metal coins, comprising: The acquisition module is used to acquire the rusted area, SEM morphology, and EDS elemental composition and distribution. A memory for storing programs for identifying counterfeit methods of the ancient metal coins; The processor and the program in the memory can be loaded and executed by the processor to implement the method for identifying counterfeit ancient metal coins.
[0021] By adopting the above technical solution, after obtaining the SEM morphology and EDS elemental composition and distribution of the rusted area of the metal coin to be tested, the elemental characteristics, locational characteristics, and crystal growth morphology characteristics of the rust layer in the rusted area are obtained using the SEM morphology and EDS elemental composition and distribution. Based on these characteristics, the probability of identifying counterfeit metal coins is generated. This technical solution, combined with the characteristics of the rusted area, can quickly distinguish between artificially induced rust and natural corrosion rust, thereby improving the accuracy of identifying counterfeit ancient metal coins.
[0022] Thirdly, this application provides a smart terminal, which adopts the following technical solution: A smart terminal includes a memory and a processor, wherein the memory stores a computer program capable of being loaded by the processor and executing any of the methods for identifying counterfeit ancient metal coins described above.
[0023] Fourthly, this application provides a computer storage medium capable of storing corresponding programs, which facilitates improving the accuracy of counterfeit detection of ancient metal coins, and adopts the following technical solution: A computer-readable storage medium storing a computer program capable of being loaded by a processor and executing any of the aforementioned methods for authenticating ancient metal coins.
[0024] In summary, this application includes at least one of the following beneficial technical effects: 1. After obtaining the SEM morphology and EDS elemental composition and distribution of the rusted area of the metal coin to be tested, the elemental characteristics, locational characteristics, and crystal growth morphology of the rust layer in the rusted area are obtained using the SEM morphology and EDS elemental composition and distribution. Based on these characteristics, the probability of identifying counterfeit metal coins is generated. This technical solution, combined with the characteristics of the rusted area, can quickly distinguish between artificially induced rust and natural corrosion rust, and improve the accuracy of identifying counterfeit ancient metal coins. 2. The rust layer element characteristics are obtained by using EDS elemental composition and distribution. These rust layer element characteristics can determine the enrichment degree of the detected elements and generate rust layer element characteristics, so that the rust layer element characteristics can more accurately represent the situation of the detected elements in the rust area and improve the accuracy of identifying counterfeits of ancient metal coins. 3. Based on the SEM morphology, the SEM images are segmented to obtain the rust layer stacking and rust layer location characteristics. The rust layer stacking morphology characteristics are then determined based on the SEM morphology. These rust layer stacking morphology characteristics are then integrated to obtain the rust layer crystal growth morphology characteristics. This technical solution makes the rust layer location characteristics and rust layer crystal growth morphology characteristics more consistent with actual conditions, improving the accuracy of counterfeit identification of ancient metal coins. Attached Figure Description
[0025] Figure 1 This is a flowchart illustrating a method for identifying counterfeit ancient metal coins provided in an embodiment of this application.
[0026] Figure 2 This is a flowchart illustrating a method for using EDS element composition and distribution according to an embodiment of this application.
[0027] Figure 3 This is a flowchart illustrating a method for processing SEM morphology provided in an embodiment of this application.
[0028] Figure 4 This is a flowchart illustrating a method for calculating the probability of counterfeit detection provided in an embodiment of this application.
[0029] Figure 5 This is a flowchart illustrating a method for calculating feature differences provided in an embodiment of this application.
[0030] Figure 6 This is a schematic diagram of the structure of an ancient metal coin authentication system provided in an embodiment of this application. Detailed Implementation
[0031] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figures 1 to 6 The present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the application.
[0032] This application discloses a method for authenticating ancient metal coins. (Refer to...) Figure 1 The plan includes: Step S101: Identify the rusted areas on the metal coin to be tested.
[0033] The tested metal coin belongs to ancient metal coinage; for example, the tested metal coin is a square-holed round coin. The metals used in the tested metal coin include copper, iron, silver, gold, and their alloys.
[0034] The rusted area is the region on the metal coin being tested where rust products have formed. For example, when the metal coin being tested is a square-holed round coin, the rusted area can be the boundary between green rust (basic copper carbonate) and red spots (cuprous oxide).
[0035] Furthermore, if multiple corrosion areas exist on the metal coin being tested, a typical corrosion area is selected from these areas. The typical corrosion area with the largest area is taken as the corrosion area used in subsequent steps. Optionally, the typical corrosion area is a corrosion area of a preset type, for example, a corrosion area where green rust and red spots meet is selected as the typical corrosion area.
[0036] Optionally, a non-destructive cleaning operation can be performed on the metal coin before identifying the rusted areas. This non-destructive cleaning operation can be dust removal using a soft-bristled brush.
[0037] Step S102: Collect SEM images of the corroded area.
[0038] SEM (Scanning Electron Microscope) morphology refers to the images and characteristics of the microstructure of a corroded area obtained by scanning electron microscopy. SEM morphology can represent the particle distribution, porosity, and layered structure within the corroded area.
[0039] Optionally, the SEM morphology is obtained by collecting cross-sections of the corroded area.
[0040] Step S103: Determine the EDS elemental composition and distribution in the rusted area.
[0041] EDS (Energy Dispersive x-ray Spectroscopy) elemental distribution is the result of visual analysis of the elemental composition and spatial distribution of the surface or cross-section of a rusted area using an energy dispersive spectrometer.
[0042] Step S104: Based on the EDS elemental composition and distribution, obtain the rust layer elemental characteristics of the metal coin to be tested.
[0043] The elemental characteristics of rust layers indicate the distribution of elements within the rusted area. In real-world scenarios, if Cl is enriched within the rusted area, it may be harmful rust; if Se / Ni is present, it may be chemically aged.
[0044] In one optional embodiment of this application, the step of obtaining the elemental characteristics of the rust layer of the metal coin to be tested based on the EDS elemental composition and distribution may include the following sub-steps S1041 to S1045, as follows: Sub-step S1041: Set the detection element.
[0045] The elements to be detected are pre-set elements used in modern chemical aging processes. For example, the elements to be detected include, but are not limited to, Se, Ni, and Si.
[0046] Sub-step S1042: Based on the EDS elemental composition and distribution, determine whether the detection element exists in the rust area.
[0047] If the detected element is present in the rusted area, proceed to steps S1044 to S1045; If no detection element is found in the rusted area, proceed to step S1043.
[0048] Sub-step S1043: If not, use the default value to generate rust layer element features.
[0049] If no detectable element is found in the rusted area, it can be preliminarily determined that the rusted area on the tested metal coin was not obtained through modern chemical aging processes.
[0050] The default value is a preset value that technicians can adjust according to actual needs. For example, the default value is set to 1. If the rust layer element features are in vector form, then the rust layer element features generated using the default value can be (1, ..., 1), where each element in the rust layer element features has a value of 1.
[0051] Sub-step S1044: If so, determine the location distribution of the detected elements in the rusted area.
[0052] If the detected elements are present in the rusted area, it indicates that the tested metal coin may have undergone modern chemical aging, and further judgment is needed.
[0053] Location distribution refers to the position of the detected element within the rusted area.
[0054] For example, the position coordinates of the detected elements are determined in the SEM topography to obtain the position distribution.
[0055] Sub-step S1045: Calculate the enrichment degree of the detected elements based on their location distribution to obtain the elemental characteristics of the rust layer.
[0056] Optionally, cluster analysis is performed on the detected elements based on their location distribution to obtain the enrichment regions corresponding to the detected elements. The total number of particles of the detected elements in each enrichment region is obtained. The mean of the total number of particles is calculated to obtain the mean content of the detected elements. Feature extraction is performed on the mean content of the detected elements to obtain the elemental characteristics of the rust layer.
[0057] The algorithms used in cluster analysis can be K-means, DBSCAN, spectral clustering, fuzzy clustering, etc.
[0058] Feature extraction is used to extract discriminative, low-redundancy features from the mean content of detected elements, ensuring that the rust layer element features are representative of the mean content of detected elements. This feature extraction can be performed using a pre-trained CNN model.
[0059] Step S105: Segment and classify the SEM morphology to obtain the location characteristics of the rust layer and the crystal growth morphology characteristics of the rust layer of the metal coin to be tested.
[0060] The locational characteristics of rust layers indicate the arrangement order of rust layers within the rusted area.
[0061] The morphological characteristics of rust layer crystal growth indicate the crystal shape and distribution of crystal particles in the rust layer within the corrosion area.
[0062] In one optional embodiment of this application, the step of segmenting and classifying the SEM morphology to obtain the locational characteristics and crystal growth morphology characteristics of the rust layer of the tested metal coin may include the following sub-steps S1051 to S1053, as follows: Sub-step S1051: Based on the SEM morphology, the SEM morphology is segmented to obtain the rust layer stacks and rust layer location features. The rust layer location features are used to describe the stacking sequence of the rust layer stacks starting from the surface of the metal coin to be tested.
[0063] Rust layering refers to the layering of different rust types within a rusted area. For example, a sorting direction is set that is perpendicular to the surface of the metal coin being tested and points away from the coin.
[0064] For example, in the rusted area of the tested metal coin, rust layers including cinnabar rust, green rust, and blue rust were detected.
[0065] For example, after obtaining the SEM morphology, the microstructure at different locations within the SEM morphology is read. The pixels in the SEM morphology are grouped according to the microstructure to obtain microstructure groups. Based on these microstructure groups, the rust layer stacks and rust layer location features are obtained.
[0066] Sub-step S1052: Based on the SEM morphology, determine the morphological characteristics of the rust layer stacks. The morphological characteristics of the rust layer stacks are used to describe the morphology of a single rust layer stack.
[0067] Optionally, for any rust layer stack in the corroded area, the morphological image of that rust layer stack in SEM morphology is taken. A preset CNN model is called to identify the rust layers in the morphological image to obtain the morphological features of the rust layer stack. For example, the CNN model identifies the crystals in the rust layer stack as dense cubic crystals.
[0068] Sub-step S1053: Integrate the morphological features of the rust layer stack to obtain the morphological features of the rust layer crystal growth.
[0069] For example, when the morphological features of the rust layer stack are represented in vector form, the morphological features of the rust layer stack are integrated, and the resulting morphological features of the rust layer crystal growth are represented in matrix form.
[0070] Step S106: Based on the elemental characteristics, locational characteristics, and crystal growth morphology of the rust layer, generate the probability of identifying counterfeit metal coins to be tested.
[0071] Optionally, if the probability of detecting a counterfeit is less than a preset probability threshold, a first-type processing suggestion is generated. If the probability of detecting a counterfeit is greater than the preset probability threshold, the type of rust in the rusted area is determined based on the elemental characteristics and crystal growth morphology of the rust layer. A second-type processing suggestion is then generated based on the type of rust.
[0072] The preset probability threshold is a preset empirical value, and technicians can adjust the specific value of the preset probability threshold according to actual needs.
[0073] If the probability of detecting counterfeit is less than the preset probability threshold, it indicates that the tested metal coin is counterfeit. The first type of handling recommendations are used to deal with counterfeit coins. For example, the first type of handling recommendations include reporting to the relevant departments, conducting secondary identification, marking and retaining the coin, and harmless disposal.
[0074] If the probability of detecting a counterfeit is greater than a preset probability threshold, it indicates that the tested metal coin is genuine. The second type of processing suggestion is related to the type of corrosion. The type of corrosion refers to the type of corrosion products within the corrosion area.
[0075] For example, when the corrosion type is cuprous oxide or basic copper carbonate, the corrosion area on the tested metal coin has little impact on the coin itself, and the second treatment suggestion is to preserve the coin in the environment. When the corrosion type is basic copper chloride, the corrosion area on the tested metal coin is harmful to the coin, and the second treatment suggestion is to remove the corrosion area.
[0076] For example, please refer to Table 1 below, which records the classification criteria for corrosion.
[0077] Table 1. Example table of corrosion classification standards Types of rust SEM morphology EDS elemental composition and distribution True or False Relationship Cuprous oxide cubic crystal, dense Predominantly Cu / O, without Cl or Se. The bottom rust of genuine coins Basic copper carbonate columnar or needle-like crystals The main components are Cu / C / O, with a small amount of Ca. rust on the surface of genuine coins Basic copper chloride Crispy powder, porous Cu / Cl / O negative electrode Harmful rust Artificial rust induction Amorphous particles, cracks Modern elements such as Se, Ni, and Si Chemically aged fake rust By employing the above technical solution, after obtaining the SEM morphology and EDS elemental composition and distribution of the corroded area of the metal coin to be tested, the elemental characteristics, locational characteristics, and crystal growth morphology of the rust layer are obtained using the SEM morphology and EDS elemental composition and distribution. Based on these characteristics, the probability of identifying counterfeit metal coins is generated. This technical solution, combined with the characteristics of the corroded area, can quickly distinguish between artificially induced rust and natural corrosion rust, improving the accuracy of identifying counterfeit ancient metal coins. The accuracy rate of genuine coin identification using this technical solution can reach over 93%, and the counterfeit coin detection rate can reach over 88%, significantly better than the traditional ultraviolet fluorescence method (accuracy rate of approximately 70%).
[0078] In the following embodiments, the counterfeit detection probability is obtained based on three characteristics: rust elemental characteristics, rust location characteristics, and rust crystal growth morphology characteristics, making the counterfeit detection probability more accurate. Therefore, this application discloses a method for calculating the counterfeit detection probability. (Refer to...) Figure 4 The plan includes: Step S401: Obtain the first counterfeit detection probability based on the elemental characteristics of the rust layer.
[0079] For example, based on the elemental characteristics of the rust layer, the types of elements within the rusted area of the metal coin to be tested are obtained. Detection elements are set, and their content within the rusted area is obtained. Based on the types of detection elements, the basic detection probability of each type is obtained. The product of the basic detection probability and the content of each detection element is calculated to obtain the element-based counterfeit detection probability. The element-based counterfeit detection probabilities of each detection element are weighted and calculated to obtain the first counterfeit detection probability.
[0080] Step S402: Calculate the similarity between the location characteristics of the rust layer and the location characteristics of the standard rust layer to obtain the second probability of identification.
[0081] Standard rust layer location characteristics refer to preset rust layer location characteristics, whereby the order of rust layer stacking within the rusted area conforms to the natural occurrence pattern of rust layers. For example, standard rust layer location characteristics represent the order of cinnabar rust, green rust, and blue rust.
[0082] For example, the Euclidean distance between the locational features of the rust layer and the locational features of the standard rust layer is calculated. A preset processing function is then invoked to convert the Euclidean distance into a second falsification probability. The larger the Euclidean distance, the smaller the second falsification probability.
[0083] Step S403: In the morphological feature library, calculate the feature difference between the morphological features of rust layer crystal growth and those of other rust layer crystal growth to obtain the morphological difference feature.
[0084] For details of this step, please refer to [link / reference]. Figure 5 The embodiments shown are not described in detail here.
[0085] Step S404: Obtain the third falsification probability based on the minimum value of the difference in morphological features.
[0086] The minimum value of the difference in morphological features is selected. A third falsification probability corresponding to this minimum value is determined in a pre-defined probability mapping table. The probability mapping table records the mapping relationship between the minimum value of the difference in morphological features and the third falsification probability.
[0087] Step S405: Calculate the first, second, and third falsification probabilities using weighted averages to obtain the falsification probability.
[0088] The weight values used in the weighted calculation can be preset. For example, the weight value for the first false positive probability is 20%, the weight value for the second false positive probability is 40%, and the weight value for the third false positive probability is 40%.
[0089] In some embodiments, if the first counterfeit detection probability is less than a first probability threshold, or the second counterfeit detection probability is less than a second probability threshold, or the third counterfeit detection probability is less than a third probability threshold, then the counterfeit detection probability will be directly set to 0. When the aforementioned situations occur, it indicates that the tested metal coin has been clearly counterfeited.
[0090] By employing the above technical solution, based on the elemental characteristics, locational characteristics, and crystal growth morphology of the rust layer, a first, second, and third false detection probability are generated sequentially. These probabilities are then weighted and calculated to obtain the final false detection probability. This technical solution makes the false detection probability more accurate.
[0091] In the following embodiments, the morphological feature library includes a genuine coin morphological feature library and a counterfeit coin morphological feature library. Different morphological feature libraries can be used to identify the morphological features of rust layer crystal growth to ensure the accuracy of the feature difference value. Therefore, this application discloses a method for calculating the feature difference value. (Refer to...) Figure 4 The plan includes: Step S501: Extract genuine coin appearance features from the genuine coin appearance feature library.
[0092] The genuine coin morphology feature library is used to store the morphological features of corrosion on genuine coins. For example, the genuine coin morphology feature library stores the morphological features corresponding to the dense crystals of naturally formed basic copper carbonate and the layered structure of cuprous oxide.
[0093] Genuine coin appearance features are any feature in the genuine coin appearance feature library.
[0094] Step S502: Calculate the difference between the morphological characteristics of the rust layer crystal growth and the morphological characteristics of the genuine coin to obtain the morphological characteristic difference of the genuine coin.
[0095] For example, the Euclidean distance between the rust layer crystal growth morphology features and each genuine coin morphology feature in the genuine coin morphology feature library is calculated to obtain a set of genuine coin morphology feature differences. The mean value in the set of genuine coin morphology feature differences is taken as the genuine coin morphology feature difference output in this step.
[0096] Step S503: Extract counterfeit coin appearance features from the counterfeit coin appearance feature library.
[0097] The counterfeit coin morphology feature library is used to store the rust morphology features of counterfeit coins. For example, the counterfeit coin morphology feature library stores the morphology features corresponding to artificially induced loose powdery rust, Se / Ni and other modern chemical aging elements.
[0098] The morphological features of counterfeit currency are any feature in the counterfeit currency morphological feature library.
[0099] Step S504: Calculate the difference between the morphological characteristics of the rust layer crystal growth and the morphological characteristics of the counterfeit coin to obtain the morphological characteristic difference of the counterfeit coin.
[0100] For example, the Euclidean distance between the morphological features of rust layer crystal growth and each counterfeit coin morphological feature in the counterfeit coin morphological feature library is calculated to obtain a set of counterfeit coin morphological feature differences. The mean value in the set of counterfeit coin morphological feature differences is taken as the counterfeit coin morphological feature difference output in this step.
[0101] Step S505: Take the larger value between the difference in appearance features of genuine coins and the difference in appearance features of counterfeit coins as the feature difference value.
[0102] By adopting the above technical solution, the feature difference values for genuine and counterfeit coins are obtained based on the genuine coin feature library and the counterfeit coin feature library, respectively. A weighted calculation of these two feature difference values generates a more accurate feature difference.
[0103] Based on the same inventive concept, embodiments of this application provide a system for authenticating ancient metal coins, including: The acquisition module 601 is used to acquire the rusted area, SEM morphology, and EDS elemental composition and distribution. The memory 602 is used to store the program for the above-mentioned method of identifying counterfeits of ancient metal coins; The processor 603 can load and execute the program in the memory to implement the above-mentioned method for identifying counterfeit ancient metal coins.
[0104] By adopting the above technical solution, after obtaining the SEM morphology and EDS elemental composition and distribution of the rusted area of the metal coin to be tested, the elemental characteristics, locational characteristics, and crystal growth morphology characteristics of the rust layer in the rusted area are obtained using the SEM morphology and EDS elemental composition and distribution. Based on these characteristics, the probability of identifying counterfeit metal coins is generated. This technical solution, combined with the characteristics of the rusted area, can quickly distinguish between artificially induced rust and natural corrosion rust, thereby improving the accuracy of identifying counterfeit ancient metal coins.
[0105] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
[0106] This application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed to identify counterfeit ancient metal coins.
[0107] Computer storage media include, for example, USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media that can store program code.
[0108] Based on the same inventive concept, this application provides a smart terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed to identify the counterfeit of ancient metal coins.
[0109] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
[0110] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.
Claims
1. A method for authenticating ancient metal coins, characterized in that, The method includes: Identify the rusted areas on the metal coin to be tested; The scanning electron microscope (SEM) morphology of the corroded area was collected; The elemental composition and distribution of the corroded area were determined by energy dispersive spectroscopy (EDS). Based on the elemental composition and distribution of EDS, the elemental characteristics of the rust layer of the metal coin under test were obtained; The SEM morphology is segmented and classified to obtain the locational characteristics and crystal growth morphology of the rust layer of the metal coin under test. Based on the elemental characteristics of the rust layer, the locational characteristics of the rust layer, and the morphological characteristics of the rust layer crystal growth, the probability of identifying counterfeit metal coins to be tested is generated.
2. The method for authenticating ancient metal coins according to claim 1, characterized in that, The step of obtaining the elemental characteristics of the rust layer of the tested metal coin based on EDS elemental composition and distribution includes: Set the detection element; Based on the composition and distribution of the EDS elements, determine whether the detected elements are present in the rusted area; If not, the default value will be used to generate the rust layer element features; If so, then determine the positional distribution of the detected element in the corroded area; Based on the location distribution, the enrichment degree of the detected elements is calculated to obtain the elemental characteristics of the rust layer.
3. The method for authenticating ancient metal coins according to claim 1, characterized in that, The step of calculating the enrichment degree of the detected elements based on the location distribution to obtain the elemental characteristics of the rust layer includes: Based on the location distribution, cluster analysis is performed on the detected elements to obtain the enriched regions corresponding to the detected elements; Obtain the total number of particles of the detected element in each enrichment region; Calculate the average total number of particles to obtain the average content of the detected elements; Feature extraction is performed on the average content of the detected elements to obtain the elemental characteristics of the rust layer.
4. The method for authenticating ancient metal coins according to claim 1, characterized in that, The process of segmenting and classifying the SEM morphology to obtain the locational characteristics and crystal growth morphology of the rust layer of the tested metal coin includes: Based on the rust crystal growth characteristics of the SEM morphology, the SEM morphology is segmented to obtain rust layer stacks and rust layer location features. The rust layer location features are used to describe the stacking order of rust layer stacks arranged starting from the surface of the metal coin under test. Based on the SEM morphology, the morphological features of the rust layer stacks are determined, and the morphological features of the rust layer stacks are used to describe the morphology of a single rust layer stack. By integrating the morphological features of the rust layer stacks, the morphological features of the rust layer crystal growth are obtained.
5. The method for authenticating ancient metal coins according to claim 1, characterized in that, The step of generating the counterfeit detection probability of the metal coin under test based on the elemental characteristics, locational characteristics, and crystal growth morphology characteristics of the rust layer includes: Based on the elemental characteristics of the rust layer, the first probability of fake detection is obtained; The similarity between the locational characteristics of the rust layer and the distribution characteristics of naturally grown rust is calculated to obtain the second probability of fake detection. In the morphological feature library, the feature difference between the morphological feature of the rust layer crystal growth and other morphological features of the rust layer crystal growth is calculated to obtain the morphological difference feature; The third spoofing probability is obtained based on the minimum value of the difference in the morphological features. The first false detection probability, the second false detection probability, and the third false detection probability are weighted and calculated to obtain the false detection probability.
6. The method for authenticating ancient metal coins according to claim 5, characterized in that, The morphological feature library includes a genuine coin morphological feature library and a counterfeit coin morphological feature library; The step of calculating the feature difference between the rust layer crystal growth morphology feature and other rust layer crystal growth morphology features in the morphology feature library includes: Take the morphological features of genuine coins from the aforementioned genuine coin morphological feature library; The difference between the morphological characteristics of the rust layer crystal growth and the morphological characteristics of the genuine coin is calculated to obtain the morphological characteristic difference of the genuine coin. Take the counterfeit currency morphology features from the counterfeit currency morphology feature library; The difference between the morphological characteristics of the rust layer crystal growth and the morphological characteristics of the counterfeit coin is calculated to obtain the morphological characteristic difference of the counterfeit coin. The larger value between the difference in the morphological features of the genuine coin and the difference in the morphological features of the counterfeit coin is taken as the feature difference value.
7. The method for authenticating ancient metal coins according to claim 1, characterized in that, The method further includes: If the probability of detecting a fake product is less than a preset probability threshold, a first type of processing suggestion is generated. When the probability of detecting counterfeits is greater than the preset probability threshold, the type of rust in the rusted area is obtained based on the elemental characteristics of the rust layer and the crystal growth morphology characteristics of the rust layer. A second type of treatment recommendation is generated based on the type of corrosion described.
8. A system for authenticating ancient metal coins, characterized in that, The system is used to perform the method for identifying counterfeit ancient metal coins as described in any one of claims 1 to 7, including: The acquisition module is used to acquire the rusted area, SEM morphology, and EDS elemental composition and distribution. A memory for storing programs for identifying counterfeit methods of the ancient metal coins; The processor and the program in the memory can be loaded and executed by the processor to implement the method for identifying counterfeit ancient metal coins.
9. A smart terminal, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any one of claims 1 to 7 for the method of identifying counterfeit ancient metal coins.
10. A computer-readable storage medium, characterized in that, The computer program stores a method for identifying counterfeit ancient metal coins as described in any one of claims 1 to 7 that can be loaded by a processor.