KR102453303B1 - Method, system and computer-readable medium of discrimiating diamond using deep learning model - Google Patents

Abstract

The present invention relates to a diamond discrimination method, system, and computer-readable medium using a deep learning learning model, and more particularly, an image obtained by irradiating ultraviolet and visible light to a diamond to be identified distributed from a diamond supplier. It relates to a diamond discrimination method, system, and computer-readable medium using a deep learning learning model, capable of discriminating whether a diamond to be discriminated is synthesized by inputting .

Description

Method, system, and computer-readable medium of diamond discrimination using deep learning learning model {METHOD, SYSTEM AND COMPUTER-READABLE MEDIUM OF DISCRIMIATING DIAMOND USING DEEP LEARNING MODEL}

The present invention relates to a diamond discrimination method, system, and computer-readable medium using a deep learning learning model, and more particularly, an image obtained by irradiating ultraviolet and visible light to a diamond to be identified distributed from a diamond supplier. It relates to a diamond discrimination method, system, and computer-readable medium using a deep learning learning model, capable of discriminating whether a diamond to be discriminated is synthesized by inputting .

The global diamond market is very large, with active distribution and transactions in each country such as the United States, Japan, and China. It is an industry.

Meanwhile, in Korea, considering the number of marriages and the gift market, at least 300,000 diamonds are in circulation, and the largest diamond among them is a natural colorless diamond mined and cut from nature. It is known that the conditions for the formation of natural diamonds are usually high temperature of about 1500℃ and high pressure of 65,000kg per square centimeter for carbon atoms at a depth of about 140 ~ 200km underground. Since natural diamonds are produced under these conditions, despite the high demand, the demand for natural diamonds cannot be satisfied due to the rarity of their natural occurrence. A method of synthesizing diamonds has been developed to meet such demand for diamonds, and with the recent development of technology, synthetic diamonds with the same physical and chemical properties as natural diamonds can be mass-produced and distributed, making synthetic diamonds similar to natural diamonds. Consumers and distributors are being harmed by malicious attempts to profit from distribution. In order to prevent this, it is necessary to make an effort to accurately identify a large amount of diamonds in a short time.

On the other hand, Patent Document 1 discloses a technique for discriminating natural diamonds and synthetic diamonds by measuring diamonds with a spectrometer and analyzing the characteristic peaks of the spectrum, but this has a problem in that it is difficult to distinguish a large number of diamonds in a short time. .

On the other hand, artificial intelligence technology is a technology that allows computers to think on their own like humans and make judgments like humans. Numerous technologies are being developed in this regard by learning the computer for a large number of specified data to solve problems. . Considering that more than 300,000 diamonds are being distributed every year in Korea, it is possible to accurately and quickly identify a large amount of diamonds using a machine learning model that learns UV images and visible light images that show the characteristics of natural and synthetic diamonds. There was a need for skills to be able to.

Registered Patent 10-2007-0076410 (diamond identification method)

The present invention is a diamond discrimination method using a deep learning learning model, which can discriminate whether the diamond is synthesized by inputting an image obtained by irradiating ultraviolet and visible light to a diamond distributed from a diamond supplier into a machine-learned discrimination model. , a system, and a computer-readable medium.

In order to solve the above problems, the present invention provides a synthetic diamond discrimination method using a deep learning learning model, comprising: irradiating UV rays to a diamond to be discriminated and acquiring a photographed UV image; inputting input data including the UV image into a machine-learning discrimination model; and outputting a discrimination result for a diamond to be discriminated from in the machine-learned discriminant model, wherein the UV image is an image obtained by irradiating light in a UV region having a wavelength of 210 nm or less in a dark room, deep learning A synthetic diamond discrimination method using a learning model is provided.

In an embodiment of the present invention, the machine-learned discrimination model may be learned by learning data including a UV image of a natural diamond and a UV image of a synthetic diamond.

In an embodiment of the present invention, the UV image of the synthetic diamond may include a UV image of the synthetic diamond using a high-temperature and high-pressure method and a UV image of the synthetic diamond using a chemical vapor deposition method.

In one embodiment of the present invention, the machine-learned discrimination model includes a plurality of individual discrimination models for each diamond supplier, and the individual discrimination model is individually based on UV images of natural diamonds and synthetic diamonds for each supplier of diamonds. is learned, and the input data may include supplier information of the corresponding discrimination target diamond.

In an embodiment of the present invention, the input data may further include a visible light image, which is an image obtained by irradiating visible light in a dark room.

In an embodiment of the present invention, the machine-learned discrimination model is learned by the visible ray image of natural diamond, the UV image and the visible ray image of the synthetic diamond, and the UV image, and in the machine learning discrimination model, the The outputting of the discrimination result may include outputting the discrimination result based on input data including the UV image and the visible ray image of the diamond to be discriminated.

In an embodiment of the present invention, the machine-learned discrimination model includes a plurality of individual discrimination models for each diamond supplier, and the individual discrimination model includes UV images and visible light images of natural diamonds for each supplier of diamonds, and synthetic diamonds for each supplier. is individually learned by the UV image and the visible ray image, and the input data may include supplier information of the corresponding discrimination target diamond.

In one embodiment of the present invention, the step of acquiring the UV image comprises: disposing a plurality of diamonds to be identified on a substrate array; obtaining a full UV image and a total visible light image by irradiating UV light and visible light to the arranged plurality of identification target diamonds; and dividing the entire visible ray image and the entire UV image into regions of each diamond to be discriminated to obtain a visible ray image and a UV image for each diamond to be discriminated; including, in the machine-learned discrimination model In the step of outputting the discrimination result for the discrimination target diamond, the discrimination result for each discrimination target diamond may be output based on the visible light image and the UV image for each discrimination target diamond.

In order to solve the above problems, an embodiment of the present invention is a synthetic diamond discrimination system using a deep learning learning model implemented with a computing device including one or more processors and one or more memories, UV for the diamond to be identified a discrimination target image acquisition unit for acquiring a UV image photographed by irradiating light; a discrimination target image input unit for inputting input data including the UV image to a machine-learned discrimination model; and a discrimination result output unit for outputting a discrimination result for a diamond to be discriminated in the machine-learned discriminant model, wherein the UV image is an image obtained by irradiating light in a UV region having a wavelength of 210 nm or less in a dark room. , provides a synthetic diamond discrimination system using a deep learning learning model.

In order to solve the above problems, an embodiment of the present invention is for implementing a synthetic diamond discrimination method using a deep learning learning model performed in a computing device including one or more processors and one or more memories, computer-readable As a medium, the computer-readable medium stores instructions for causing a computing device to perform the following steps, the steps comprising: obtaining a photographed UV image by irradiating UV rays to a diamond to be discriminated; inputting input data including the UV image into a machine-learning discrimination model; and outputting a discrimination result for a diamond to be discriminated from in the machine-learned discriminant model, wherein the UV image is an image obtained by irradiating light in a UV region having a wavelength of 210 nm or less in a dark room, computer- A readable medium is provided.

According to an embodiment of the present invention, it is possible to exhibit the effect of quickly and accurately discriminating whether a large amount of diamonds are natural or synthetic using a deep learning learning model.

According to an embodiment of the present invention, it is possible to exert the effect of deriving a discrimination result as to whether or not diamond is synthesized for each diamond supplier in consideration of the characteristics according to the diamond processing method by the machine-learned discrimination model.

According to an embodiment of the present invention, through a deep learning learning model that can rapidly discriminate diamonds in a non-destructive way without destroying them, it is possible to exert the effect of determining whether natural diamonds and synthetic diamonds are provided by each supplier of diamonds. .

According to an embodiment of the present invention, by using not only the UV image of the diamond to be discriminated but also the visible ray image as input data, it is possible to determine whether the diamond is synthesized by linking the UV image and the visible ray image to obtain a more accurate discrimination result. derivable effects can be achieved.

According to an embodiment of the present invention, through processing of a UV image and a visible light image obtained by arranging a plurality of diamonds on a substrate array, it is possible to more efficiently and quickly derive a discrimination result on whether a large number of diamonds are synthesized at the same time. possible effect can be exerted.

1 schematically shows the overall diamond supply form in the diamond distribution market.
2 schematically shows the operation of a synthetic diamond discrimination system according to an embodiment of the present invention.
3 schematically shows the internal configuration of a synthetic diamond discrimination system using a deep learning learning model according to an embodiment of the present invention.
4 schematically shows the steps of performing a diamond discrimination method using a deep learning learning model according to an embodiment of the present invention.
5 schematically shows a UV image of a natural diamond and a UV image of a synthetic diamond according to an embodiment of the present invention.
6 schematically shows a machine-learned discriminant model according to an embodiment of the present invention.
7 schematically shows a UV image of a synthetic diamond according to an embodiment of the present invention.
8 schematically shows visible light images and UV images of natural diamonds and synthetic diamonds according to an embodiment of the present invention.
9 schematically shows detailed steps of the step of acquiring a UV image according to an embodiment of the present invention.
10 schematically illustrates an operation of a discrimination target image acquisition unit according to an embodiment of the present invention.
11 schematically shows visible light images and UV images of a plurality of diamonds to be discriminated according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. However, the present invention may be embodied in several different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part is "connected" with another part, this includes not only the case of being "directly connected" but also the case of being "electrically connected" with another element interposed therebetween. . Also, when a part "includes" a certain component, it means that other components may be further included, rather than excluding other components, unless otherwise stated.

Also, terms including an ordinal number such as first, second, etc. may be used to describe various elements, but the elements are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component. and/or includes a combination of a plurality of related listed items or any of a plurality of related listed items.

In this specification, a "part" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. In addition, one unit may be implemented using two or more hardware, and two or more units may be implemented by one hardware. Meanwhile, '~ unit' is not limited to software or hardware, and '~ unit' may be configured to be in an addressable storage medium or may be configured to reproduce one or more processors. Thus, as an example, '~' denotes components such as software components, object-oriented software components, class components, and task components, and processes, functions, properties, and procedures. , subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. The functions provided in the components and '~ units' may be combined into a smaller number of components and '~ units' or further separated into additional components and '~ units'. In addition, components and '~ units' may be implemented to play one or more CPUs in a device or secure multimedia card.

The "user terminal" referred to below may be implemented as a computer or portable terminal that can access a server or other terminal through a network. Here, the computer includes, for example, a laptop equipped with a web browser (WEB Browser), a desktop, a laptop, and the like, and the portable terminal is, for example, a wireless communication device that guarantees portability and mobility. , PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code) Division Multiple Access)-2000, W-Code Division Multiple Access (W-CDMA), and Wibro (Wireless Broadband Internet) terminals may include all types of handheld-based wireless communication devices. In addition, "network" refers to a wired network such as a local area network (LAN), a wide area network (WAN) or a value added network (VAN), or a mobile radio communication network or satellite. It may be implemented in any kind of wireless network, such as a communication network.

1 schematically shows the overall diamond supply form in the diamond distribution market.

As shown in FIG. 1 , the diamond distribution market may include a plurality of diamond suppliers, a diamond distributor that receives and distributes diamonds from the diamond supplier, and a plurality of retailers that purchase or sell diamonds from the distributors. Diamonds are generally imported into Korea by importers, and then delivered to consumers such as retailers through intermediate distributors. Afterwards, consumers may perform second-hand transactions between consumers or resell to distributors again. On the other hand, a significant number of diamonds are traded annually in Korea, and the synthetic diamond identification system according to the present invention is a diamond distributor due to malicious attempts by the supplier of diamonds to obtain profits by distributing synthetic diamonds like natural diamonds. A large amount of diamonds supplied from suppliers can be identified quickly and accurately in order to prevent damage to them.

2 schematically shows the operation of the synthetic diamond discrimination system 1000 according to an embodiment of the present invention.

According to an embodiment of the present invention, as shown in FIG. 2A , the synthetic diamond discrimination system 1000 discriminates a diamond to be discriminated based on input data including a UV image of the diamond to be discriminated. results can be derived.

Specifically, the input data includes a UV image, which is an image obtained by irradiating light in a UV region having a wavelength of 210 nm or less in a dark room.

The synthetic diamond discrimination system 1000 may derive a discrimination result for a diamond to be discriminated based on such input data. By inputting the input data including the UV image of the diamond to be differentiated into the machine-learned discrimination model, it is possible to derive the discrimination result of discriminating whether the diamond is a synthetic diamond or a natural diamond.

The discrimination results derived from the machine-learned discrimination model include the results of determining whether natural diamonds and synthetic diamonds are, and in the case of synthetic diamonds, synthetic diamonds using a high-temperature and high-pressure method according to the manufacturing process and chemical vapor deposition method It can be determined whether it is a synthetic diamond used. Preferably, the synthetic diamond discrimination system 1000 may determine whether natural diamonds and synthetic diamonds exist for each diamond supplier, and may derive a discrimination result based thereon.

Methods for synthesizing diamond include High Pressure-High Temperature (HPHT) and Chemical Vapor Deposition (CVD). In synthetic diamonds, nitrogen is mostly removed due to additives added during the manufacturing process, and due to the difference in nitrogen content of natural diamonds, differences in nitrogen fluorescence and absorption occur when irradiated with UV (Ultra Violet) to diamonds. Due to these factors, it is possible to easily distinguish between natural diamonds and synthetic diamonds. The machine-learned discrimination model to derive the discrimination result based on the input data is learned based on the learning data including the UV images of a number of previously discriminated diamonds, and derives the discrimination result of the UV image of the diamond to be discriminated. can do. Preferably, the machine-learned discrimination model is learned by learning data including a UV image of a natural diamond and a UV image of a synthetic diamond, and the UV image of the synthetic diamond is a UV image of a synthetic diamond using a high-temperature and high-pressure method, and Includes UV images of synthetic diamonds using chemical vapor deposition.

Meanwhile, as shown in (b) of FIG. 2 , the input data may further include a visible light image, which is an image obtained by irradiating visible light to a diamond to be identified. The machine-learned discrimination model to derive the discrimination result based on the input data is learned based on learning data including UV images and visible light images of a plurality of diamonds discriminated and accumulated in the past, and A discrimination result can be derived based on the UV image and the visible light image. In addition, by learning the relationship between the UV image and the visible light image, the discrimination result of the corresponding diamond can be derived. Preferably, the discrimination result is at least one of natural diamond and synthetic diamond, whether it is a synthetic diamond using a high-temperature and high-pressure method, and a synthetic diamond using a chemical vapor deposition method, supplier information of the corresponding discrimination target diamond, and the process method used. may include

The synthetic diamond discrimination system 1000 shown in FIG. 2 may further include elements other than the elements shown, but for convenience, only the elements related to the synthetic diamond discrimination system 1000 according to embodiments of the present invention are displayed. did.

3 schematically shows a system for discriminating a synthetic diamond using a deep learning learning model according to an embodiment of the present invention, and FIG. 4 is a method for discriminating a diamond using a deep learning learning model according to an embodiment of the present invention. The execution steps are schematically shown.

The synthetic diamond discrimination system 1000 according to the above embodiment may be implemented by a computing device using one or more processors and one or more memories.

Such a computing device may include a processor (A), a bus (corresponding to a double-headed arrow between a processor, a memory, and a network interface), a network interface (B), and a memory (C). The memory C may store learning data used in the discrimination model of the present invention as learning data learned in implementing the OS (operating system) and the machine-learned discrimination model. Alternatively, the learning data may refer to modeling information in which deep learning has been performed. In the processor A, the discrimination target image acquisition unit 1100 , the discrimination target image input unit 1200 , and the discrimination result output unit 1300 may be executed. In other embodiments, the synthetic diamond discrimination system 1000 may include more components than those of FIG. 3 .

The memory is a computer-readable recording medium and may include a random access memory (RAM), a read only memory (ROM), and a permanent mass storage device such as a disk drive. These software components may be loaded from a computer-readable recording medium separate from the memory using a drive mechanism (not shown). The separate computer-readable recording medium may include a computer-readable recording medium (not shown) such as a floppy drive, a disk, a tape, a DVD/CD-ROM drive, and a memory card. In another embodiment, the software components may be loaded into the memory through the network interface (B) rather than the computer-readable recording medium.

A bus may enable communication and data transfer between components of a computing device. The bus may be configured using a high-speed serial bus, a parallel bus, a storage area network (SAN), and/or other suitable communication technology.

The network interface B may be a computer hardware component for connecting the computing device of the synthetic diamond discrimination system 1000 to a computer network. The network interface (B) may connect the emotion recognition system to a computer network through a wireless or wired connection.

The processor A may be configured to process instructions of a computer program by performing input/output operations implementing the basic arithmetic, logic, and synthetic diamond discrimination system 1000 . Instructions may be provided to the processor by the memory (C) or network interface (B) and via a bus. The processor may be configured to execute program codes for the discrimination target image acquisition unit 1100 , the discrimination target image input unit 1200 , and the discrimination result output unit 1300 . Such program code may be stored in a recording device such as a memory.

The discrimination target image acquisition unit 1100 , the discrimination target image input unit 1200 , and the discrimination result output unit 1300 may be configured to perform a synthetic diamond discrimination method to be described below. In the above-described processor, some components may be omitted, additional components not shown may be further included, or two or more components may be combined according to a discrimination method.

On the other hand, such a computing device preferably corresponds to a personal computer or server, and in some cases, a smart phone, a tablet, a mobile phone, a video phone, and an e-book reader (e). -book reader), desktop PC, laptop PC, netbook PC, personal digital assistant (PDA, hereinafter referred to as 'PDA') and portable A portable multimedia player (PMP, hereinafter referred to as 'PMP'), an mp3 player, a mobile medical device, a camera, and a wearable device (eg, a head-mounted device) A head-mounted device (HMD, for example, will be referred to as 'HMD'), electronic clothing, electronic bracelets, electronic necklaces, electronic accessories, electronic tattoos, or smart watches ) and so on.

The synthetic diamond discrimination system 1000 using the deep learning learning model of the present invention includes: a discrimination target image acquisition unit 1100 for acquiring a UV image photographed by irradiating UV rays with respect to the discrimination target diamond; a discrimination target image input unit 1200 for inputting input data including the UV image to a machine-learned discrimination model; and a discrimination result output unit 1300 for outputting a discrimination result for a diamond to be discriminated in the machine-learned discrimination model.

Specifically, the discrimination target image acquisition unit 1100 acquires a photographed UV image by irradiating UV rays to the discrimination target diamond. Preferably, the UV image is an image obtained by irradiating light in the UV region having a wavelength of 210 nm or less in a dark room. Due to the difference in the manufacturing process of natural diamonds and synthetic diamonds, when a diamond to be differentiated is irradiated with light in the UV region, the UV image may show a difference in color due to the difference in the absorption rate of nitrogen contained in each diamond. In the machine-learned discrimination model, learning of input data may be performed based on the feature information including the color difference.

The discrimination target image input unit 1200 inputs input data including the UV image acquired by the discrimination target image acquisition unit 1100 into the machine-learned discrimination model.

The discrimination result output unit 1300 may output a discrimination result with respect to the discrimination target diamond in the machine-learned discrimination model. Preferably, the machine-learned discrimination model may be learned by learning data including a UV image of a natural diamond and a UV image of a synthetic diamond. The UV image of the synthetic diamond includes a UV image of the synthetic diamond using a high-temperature and high-pressure method and a UV image of the synthetic diamond using a chemical vapor deposition method.

In this way, in the present invention, it is possible to determine whether the diamond to be identified is synthesized by inputting a UV image obtained by irradiating UV rays to the diamond to be identified in the machine-learned discrimination model, and the type of synthetic diamond according to the manufacturing process It can also be determined whether

Hereinafter, the criteria for determining whether natural diamonds and synthetic diamonds are synthesized will be described in more detail.

5 schematically shows a UV image of a natural diamond and a UV image of a synthetic diamond according to an embodiment of the present invention.

Methods for synthesizing diamond include High Pressure-High Temperature (HPHT) and Chemical Vapor Deposition (CVD). The high-temperature and high-pressure method uses the principle of generating natural diamonds to form a diamond phase while maintaining the graphite raw material in a high-temperature and high-pressure environment for a preset period. By removing the nitrogen compound, a transparent colorless single crystal can be produced. The chemical vapor deposition method is a method of growing single crystal diamond using methane gas and hydrogen gas under high-density plasma using microwaves, and has little nitrogen incorporation. In this way, nitrogen by the high-temperature and high-pressure method is removed by adding boron, while diamond by chemical vapor deposition has almost no nitrogen, whereas natural diamond contains more than 99% nitrogen. When a diamond is irradiated with UV (Ultra Violet), there is a difference in the fluorescence and absorption of nitrogen. Due to these characteristics, it is possible to easily distinguish between natural diamonds and synthetic diamonds. In order to derive the discrimination result based on the input data, the machine-learned discrimination model is learned based on the learning data including the UV images of a number of previously discriminated diamonds, and the discrimination result of the UV images of the diamond to be discriminated against. can be derived

Fig. 5 (a) shows a UV image of a natural diamond, and Fig. 5 (b) shows a UV image of a synthetic diamond. As shown in FIG. 5 , it is shown that natural diamonds clearly show a uniform fluorescence reaction of blue color, and synthetic diamonds exhibit red fluorescence depending on the presence or absence of a nitrogen component of diamond. As such, natural diamonds and synthetic diamonds have differences in images depending on fluorescence expression, so they can be classified into natural diamonds and synthetic diamonds by learning them on a machine-learned discriminant model. In addition, as shown in FIG. 5, natural diamonds and synthetic diamonds have an ideal cutting ratio of the table surface of the crown part of the synthetic diamond in terms of not only the difference in color but also the cutting surface, whereas the natural diamond has the weight at the time of cutting. It can be seen that the table surface was cut very narrowly for fear of reduction. It is possible to discriminate a diamond to be discriminated by such a characteristic, and preferably, the discrimination result output unit 1300 extracts the characteristic information including the color and the cutting ratio, and adds the characteristic information and the machine-learned discrimination model. Based on this, the discrimination result can be output. The machine-learned discriminant model can be learned by learning data including a UV image of a natural diamond and a UV image of a synthetic diamond, and the UV image of the synthetic diamond is a UV image of a synthetic diamond using a high-temperature and high-pressure method and chemistry It may include a UV image of the synthetic diamond using the deposition method.

6 schematically shows a machine-learned discriminant model according to an embodiment of the present invention, and FIG. 7 schematically shows a UV image of a synthetic diamond according to an embodiment of the present invention.

As described above, in the case of synthetic diamond, it can be synthesized through other manufacturing processes such as High Pressure-High Temperature (HPHT) and Chemical Vapor Deposition (CVD) method, and even if the same method is used, There are differences in the nitrogen content of diamonds due to the manufacturing process for each supplier supplying each diamond, and differences in the content of additives (eg, boron) added to remove nitrogen. As shown in FIG. 6 , in an embodiment of the present invention, the machine-learned discrimination model includes a plurality of individual discrimination models for each diamond supplier. The individual discrimination model is individually learned by the UV images of natural diamonds and the UV images of synthetic diamonds for each supplier of diamonds, and the input data may include supplier information of the corresponding discrimination target diamond. Figure 7 (a) shows the UV image of the synthetic diamond prepared through the high temperature and high pressure method of supplier A, and Figure 7 (b) shows the UV image of the synthetic diamond produced through the chemical vapor deposition method of supplier B. . The synthetic diamond of Supplier A shows a light blue fluorescence color as shown in FIG. 7(a) due to the residual boron added during the high-temperature and high-pressure process, and the synthetic diamond of Supplier B is 100 ppm in the chemical vapor deposition process. It is shown that a green color as shown in (b) of FIG. 7 appears due to the mixing of nitrogen. In addition, the synthetic diamond of Supplier A has a black carbon residue in the center. In this way, feature information including differences in nitrogen content, additive content, and residual carbon content due to differences in manufacturing processes for each diamond supplier is extracted, and the discrimination result for the diamond to be discriminated based on the feature information and individual discrimination model can be derived.

In addition, the characteristic information may include information on a cutting surface according to a cutting method for each supplier. In the present invention, by implementing an individual discrimination model for each diamond supplier for discrimination target diamonds for each supplier, each individual discrimination model can be implemented by individually learning the UV images of natural diamonds and synthetic diamonds for each supplier. The input data may include supplier information of the corresponding diamond to be discriminated, or supplier information may also be discriminated in a machine-learned discrimination model. Furthermore, each individual discrimination model may discriminate the producer of the corresponding discrimination target diamond for each supplier.

In this way, the synthetic diamond discrimination system 1000 of the present invention implements an individual discrimination model for each diamond supplier, thereby exhibiting the effect of discriminating whether a diamond is synthesized according to a different manufacturing process for each diamond supplier.

8 schematically shows visible light images and UV images of natural diamonds and synthetic diamonds according to an embodiment of the present invention.

According to an embodiment of the present invention, the input data input to the machine-learned discrimination model further includes a visible ray image, which is an image obtained by irradiating light in a visible ray region, and includes a UV image and visible ray of a diamond to be discriminated. The discrimination result is output based on the input data including the image. The machine-learned discrimination model is learned by the visible light image of natural diamond, the UV image and the visible light image of the synthetic diamond, and the UV image. (a) of FIG. 8 shows visible light images of natural diamonds, synthetic diamonds prepared by a high-temperature and high-pressure method, and synthetic diamonds prepared by chemical vapor deposition, and FIG. 8(b) is (a) of FIG. shows the UV image of each diamond. The discrimination result output unit 1300 extracts feature information according to the difference between each diamond to be identified in the visible ray image and feature information according to the difference in the UV image according to one or more angles, and the machine-learned discrimination model can output the discrimination result of the diamond to be discriminated. For example, in the case of natural diamonds, there is a characteristic of having a table in the direction of 100 faces depending on the shape of the raw stone, and there is a difference in the cutting state of the triple point and the contact point between the natural diamond and the synthetic diamond with the bezel near the table. Based on the characteristic information as described above, the relationship between the UV image and the visible ray image may be learned to derive a discrimination result of the corresponding diamond. Preferably, the machine-learned discrimination model includes a plurality of individual discrimination models for each diamond supplier, and the individual discrimination model includes UV images and visible light images of natural diamonds by suppliers of diamonds and UV images of synthetic diamonds by suppliers and It is individually learned by the visible light image, and the input data may include supplier information of the corresponding diamond to be identified.

In this way, in the present invention, a more accurate diamond discrimination result by determining whether a diamond to be discriminated is synthesized in consideration of the relationship between the visible ray image and the UV image through the discrimination model learned from the visible ray image and the UV image. can be effective in providing

9 schematically shows detailed steps of the step of acquiring a UV image according to an embodiment of the present invention, and FIG. 10 schematically shows the operation of the discrimination target image acquisition unit 1100 according to an embodiment of the present invention. and FIG. 11 schematically shows visible light images and UV images of a plurality of diamonds to be identified according to an embodiment of the present invention.

According to an embodiment of the present invention, the discrimination target image acquisition unit 1100 may perform steps S1100 to S1300 as shown in FIG. 9 .

Specifically, in step S1100, a plurality of discrimination target diamonds are arranged on a substrate array. As shown in FIG. 10A , a plurality of diamonds to be identified may be disposed on a substrate array on which the diamonds to be identified are disposed.

In step S1200, UV rays and visible rays are irradiated to the arranged plurality of identification target diamonds to obtain a full UV image and a total visible light image. Preferably, the UV rays may be irradiated in a dark room. By irradiating light in the UV region to a plurality of diamonds to be identified disposed on the substrate array, an entire UV image of the diamonds to be identified can be obtained as shown in FIG. By irradiating light, a full visible light image can also be obtained.

In step S1300, as shown in (b) of FIG. 10, by dividing the entire visible ray image and the entire UV image into regions of a diamond to be identified, a visible ray image and a UV image for each diamond to be identified are obtained. can

Thereafter, in the step (S3000) of outputting the discrimination result for the discrimination target diamond in the machine-learned discrimination model, the visible ray image and UV for each discrimination target diamond as shown in FIG. Based on the image, a discrimination result for each of the respective discrimination target diamonds may be output. Preferably, the UV image and visible ray image of each diamond to be identified derived from the total UV image and the total visible ray image are also discriminated by an individual discrimination model for each diamond supplier, so that the differentiation result according to the supplier may be derived. have.

On the other hand, FIG. 11 (a) shows a total visible ray image obtained by arranging samples of a diamond to be identified with similar quality, and FIG. 11 (b) shows an overall UV image. 11(a) and 11(b), a plurality of diamond samples can be arranged at once and proceeded simultaneously, and only four diamond samples are shown in FIG. 11, but in an embodiment of the present invention Without limiting the number of diamonds to be discriminated, the results of discrimination for diamonds to be discriminated against are obtained by acquiring all UV images and all visible light images for a large number of diamonds, such as 10, 100, and 200 or more, if necessary. can be derived

In this way, in the present invention, through processing of a UV image and a visible ray image obtained by arranging a plurality of diamonds on a substrate array, the discrimination result on whether a large number of diamonds is synthesized at the same time can be derived more efficiently and quickly. can have an effect.

An embodiment of the present invention may also be implemented in the form of a recording medium including instructions executable by a computer, such as a program module executed by a computer. Computer-readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media. In addition, computer-readable media may include both computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Communication media typically includes computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave, or other transport mechanism, and includes any information delivery media.

Although the methods and systems of the present invention have been described with reference to specific embodiments, some or all of their components or operations may be implemented using a computer system having a general purpose hardware architecture.

The foregoing description of the present invention is for illustration, and those of ordinary skill in the art to which the present invention pertains can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. For example, each component described as a single type may be implemented in a distributed manner, and likewise components described as distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

Claims (10)

As a synthetic diamond discrimination method using a deep learning learning model,
irradiating UV rays to the diamond to be identified and acquiring a photographed UV image;
inputting input data including the UV image into a machine-learning discrimination model; and
outputting a discrimination result for a diamond to be discriminated from the machine-learned discriminant model;
The UV image is an image obtained by irradiating light in the UV region having a wavelength of 210 nm or less in a dark room,
The machine-learned discrimination model includes an individual discrimination model for each diamond supplier,
The individual discrimination model is individually learned by the UV image of the natural diamond and the UV image of the synthetic diamond for each diamond supplier,
The input data is a synthetic diamond discrimination method using a deep learning learning model, including supplier information of the corresponding discrimination target diamond.
The method according to claim 1,
The machine-learned discriminant model is
A synthetic diamond discrimination method using a deep learning learning model, learned by learning data including UV images of natural diamonds and UV images of synthetic diamonds.
3. The method according to claim 2,
The UV image of the synthetic diamond is,
A synthetic diamond discrimination method using a deep learning learning model, including a UV image of a synthetic diamond using a high-temperature and high-pressure method and a UV image of a synthetic diamond using a chemical vapor deposition method.
delete delete delete As a synthetic diamond discrimination method using a deep learning learning model,
irradiating UV rays to the diamond to be identified and acquiring a photographed UV image;
inputting input data including the UV image into a machine-learning discrimination model; and
outputting a discrimination result for a diamond to be discriminated from the machine-learned discriminant model;
The UV image is an image obtained by irradiating light in the UV region having a wavelength of 210 nm or less in a dark room,
The input data is
Further comprising a visible light image, which is an image obtained by irradiating visible light in a dark room,
The machine-learned discrimination model is learned by the visible light image of natural diamond, the visible light image of the UV image and the synthetic diamond, and the UV image,
The step of outputting the discrimination result for the discrimination target diamond in the machine learning discrimination model,
outputting the discrimination result based on input data including the UV image and the visible light image of the diamond to be discriminated;
The machine-learned discrimination model includes an individual discrimination model for each diamond supplier,
The individual discrimination model is individually learned by the UV image and visible ray image of natural diamond by supplier of diamond, and UV image and visible ray image of synthetic diamond by supplier,
The input data is a synthetic diamond discrimination method using a deep learning learning model, including supplier information of the corresponding discrimination target diamond.
The method according to claim 1,
The step of acquiring the UV image is,
disposing a plurality of diamonds to be identified on a substrate array;
obtaining a full UV image and a total visible light image by irradiating UV light and visible light to the arranged plurality of identification target diamonds; and
Including; dividing the entire visible ray image and the entire UV image into regions of each diamond to be discriminated to obtain a visible ray image and a UV image for each diamond to be discriminated;
In the machine-learned discrimination model, the step of outputting the discrimination result for the discrimination target diamond is outputting the discrimination result for each discrimination target diamond based on the visible light image and the UV image for each discrimination target diamond, A synthetic diamond discrimination method using a deep learning learning model.
A synthetic diamond discrimination system using a deep learning learning model implemented as a computing device including one or more processors and one or more memories,
a discrimination target image acquisition unit for acquiring a photographed UV image by irradiating UV rays on the discrimination target diamond;
a discrimination target image input unit for inputting input data including the UV image to a machine-learned discrimination model; and
and a discrimination result output unit for outputting a discrimination result for a diamond to be discriminated from in the machine-learned discriminant model.
The UV image is an image obtained by irradiating light in the UV region having a wavelength of 210 nm or less in a dark room,
The machine-learned discrimination model includes an individual discrimination model for each diamond supplier,
The individual discrimination model is individually learned by the UV image of the natural diamond and the UV image of the synthetic diamond for each diamond supplier,
The input data is a synthetic diamond discrimination system using a deep learning learning model, including supplier information of the corresponding discrimination target diamond.
As a computer-readable medium for implementing a synthetic diamond discrimination method using a deep learning learning model performed on a computing device including one or more processors and one or more memories,
The computer-readable medium stores instructions for causing a computing device to perform the following steps:
obtaining a UV image photographed by irradiating UV rays with respect to the diamond to be discriminated;
inputting input data including the UV image into a machine-learning discrimination model; and
outputting a discrimination result for a diamond to be discriminated from the machine-learned discriminant model;
The UV image is an image obtained by irradiating light in the UV region having a wavelength of 210 nm or less in a dark room,
The machine-learned discrimination model includes an individual discrimination model for each diamond supplier,
The individual discrimination model is individually learned by the UV image of the natural diamond and the UV image of the synthetic diamond for each diamond supplier,
The input data includes information on a supplier of the corresponding discrimination target diamond, a computer-readable medium.

Images