CN113008865A - Method, device, medium and equipment for identifying jewelry jade - Google Patents
Method, device, medium and equipment for identifying jewelry jade Download PDFInfo
- Publication number
- CN113008865A CN113008865A CN202110240147.8A CN202110240147A CN113008865A CN 113008865 A CN113008865 A CN 113008865A CN 202110240147 A CN202110240147 A CN 202110240147A CN 113008865 A CN113008865 A CN 113008865A
- Authority
- CN
- China
- Prior art keywords
- similar
- species
- detected
- jewelry
- deep learning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000010977 jade Substances 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000001237 Raman spectrum Methods 0.000 claims abstract description 90
- 238000013135 deep learning Methods 0.000 claims abstract description 39
- 238000001069 Raman spectroscopy Methods 0.000 claims abstract description 33
- 238000012549 training Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 14
- 238000010801 machine learning Methods 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 8
- 241000894007 species Species 0.000 description 54
- 239000000126 substance Substances 0.000 description 42
- 239000010437 gem Substances 0.000 description 25
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 23
- 229910001751 gemstone Inorganic materials 0.000 description 22
- 239000013078 crystal Substances 0.000 description 18
- 241001503485 Mammuthus Species 0.000 description 15
- 238000010586 diagram Methods 0.000 description 14
- 239000000203 mixture Substances 0.000 description 11
- 239000010453 quartz Substances 0.000 description 10
- 235000012239 silicon dioxide Nutrition 0.000 description 10
- 238000005070 sampling Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000001514 detection method Methods 0.000 description 5
- 229910052500 inorganic mineral Inorganic materials 0.000 description 5
- 239000012491 analyte Substances 0.000 description 4
- 229940023487 dental product Drugs 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 239000011707 mineral Substances 0.000 description 4
- 235000010755 mineral Nutrition 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000001228 spectrum Methods 0.000 description 4
- 238000002329 infrared spectrum Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000013136 deep learning model Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000005284 excitation Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000000103 photoluminescence spectrum Methods 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 239000000377 silicon dioxide Substances 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 241000220317 Rosa Species 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000007789 gas Substances 0.000 description 1
- 229910010272 inorganic material Inorganic materials 0.000 description 1
- 239000011147 inorganic material Substances 0.000 description 1
- 239000003446 ligand Substances 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000009659 non-destructive testing Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000011049 pearl Substances 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 229910052611 pyroxene Inorganic materials 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000011257 shell material Substances 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 235000013619 trace mineral Nutrition 0.000 description 1
- 239000011573 trace mineral Substances 0.000 description 1
- 230000007723 transport mechanism Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/65—Raman scattering
Landscapes
- Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The article relates to a jewelry jade appraisal method based on deep learning, which is applied to a Raman spectrometer and comprises the following steps: acquiring a Raman spectrum of an object to be detected; inputting the Raman spectrum into a first-stage identification model, and determining the similar category or the species category of the object to be detected; and if the object to be detected is of a certain similar type, selecting a second-stage identification model corresponding to the similar type, inputting the Raman spectrum into the corresponding second-stage identification model, and outputting the species type of the object to be detected. Through the two-stage identification model, the similar category of the object to be detected is determined, then the second-stage identification model corresponding to the similar category is selected, the object to be detected is determined to be the specific category in the similar category, and the identification accuracy is effectively improved.
Description
Technical Field
The present invention relates to the field of identification and recognition of chemical substances, and in particular to a jewelry jade identification method, device, medium and equipment.
Background
In the related art, the spectrum detection technology mainly includes infrared spectrum analysis and laser-raman spectrum analysis, and the raman spectrometer mainly adopts a laser micro-raman spectrometer. However, the laser micro-raman spectrometer is expensive, complex in operation and inconvenient to carry, and can only detect the species of the jewelry jade in a laboratory. Handheld raman spectrometers have shown excellent and irreplaceable advantages in field identification applications, however, in the aspect of gemstone identification, handheld raman spectrometers are limited by the equipment itself, the resolution of the acquired raman spectra is low, and identification is not easy, and the accuracy of handheld raman spectrometers is greatly affected by the uncertainty of the sampling environment, the sampling personnel and the sampling normative in the detection process. In addition, the gem can generate background fluorescence of different degrees under the excitation of laser, and the interference of the fluorescence has great influence on the weak Raman scattering signal, so that the final Raman spectrum can not be imaged, and the accuracy of the detection result is influenced.
Therefore, optimizing or developing a new raman spectroscopy analysis and identification algorithm is an urgent problem to be solved for improving the identification of precious stones by the handheld raman spectrometer.
Disclosure of Invention
To overcome the problems in the related art, provided herein are a method, apparatus, medium, and device for jewelry jade evaluation based on deep learning.
According to a first aspect of the present disclosure, there is provided a method for identifying jewelry jade based on deep learning, applied to a raman spectrometer, comprising:
acquiring a Raman spectrum of an object to be detected;
inputting the Raman spectrum into a first-stage identification model, and determining the similar category or the species category of the object to be detected;
and if the object to be detected is of a certain similar type, selecting a second-stage identification model corresponding to the similar type, inputting the Raman spectrum into the corresponding second-stage identification model, and outputting the species type of the object to be detected.
The jewelry jade identification method based on deep learning further comprises the following steps: and classifying the jewelry jades of different species with similar Raman spectra into the same similar species.
The Raman spectrometer comprises: a hand-held Raman spectrometer.
The first-stage identification model is a model obtained after deep learning method training, is trained by using sample data of the jewelry jades marked with similar categories or species categories, and is used for identifying whether the object to be detected is the jewelry jades or not and determining the similar categories or species categories of the object to be detected.
The second-stage identification model comprises a plurality of models obtained after deep learning or other machine learning methods are trained, and each model is trained by using sample data of a group of jewelry jades with similar Raman spectra marked with species classes and is used for identifying the species classes of the object to be detected.
According to another aspect herein, there is provided a deep learning based jewelry jade appraisal apparatus comprising:
the Raman spectrum acquisition module is used for acquiring a Raman spectrum of the object to be detected;
the first identification module is used for inputting the Raman spectrum into a first-stage identification model and determining the similar category or the species category of the object to be detected;
the model selection module is used for selecting a second-level identification model corresponding to a similar category if the object to be detected is of the similar category;
and the second identification module is used for inputting the Raman spectrum into the corresponding second-stage identification model and outputting the species to be detected.
The first identification module is a model obtained after deep learning method training, uses sample data of the jewelry jades marked with similar categories or species categories for training, is used for identifying whether the object to be detected is the jewelry jades or not, and determines the similar categories or species categories of the object to be detected.
The second identification module comprises a plurality of models obtained after deep learning or other machine learning methods are trained, and each model is trained by using sample data of a group of jewelry jades with similar Raman spectra marked with species classes and is used for identifying the species classes of the object to be detected.
According to another aspect herein, there is provided a computer readable storage medium having stored thereon a computer program which, when executed, performs the steps of a method of authenticating a jewelry jade based on deep learning.
According to another aspect herein, there is provided a computer apparatus comprising a processor, a memory and a computer program stored on the memory, the processor when executing the computer program implementing the steps of a method of authenticating a jewelry jade based on deep learning.
The deep learning algorithm is introduced, the two-stage recognition models are trained based on the characteristics of the jewelry jade, the similar category of the object to be detected can be determined by the first-stage recognition model, the corresponding second-stage recognition model is selected for the object to be detected with the similar category according to the similar category, the specific category of the object to be detected is determined, and the recognition accuracy is effectively improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention as claimed.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the disclosure, and are incorporated in and constitute a part of this specification. In the drawings:
FIG. 1 is a flow diagram illustrating a method for deep learning-based jewelry jade qualification according to an exemplary embodiment.
Figure 2 is a schematic illustration of a raman spectrum of an ivory tooth shown in accordance with an exemplary embodiment.
Fig. 3 is a schematic raman spectrum of a mammoth ivory shown in accordance with an exemplary embodiment.
Fig. 4 is a schematic diagram of raman spectra of agate and crystal, shown in accordance with an example embodiment.
Fig. 5 is a block diagram illustrating a deep learning based jewelry jade appraisal apparatus according to an exemplary embodiment.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the drawings of the embodiments of the present invention, and it is obvious that the described embodiments are some but not all of the embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments herein without making any creative effort, shall fall within the scope of protection. It should be noted that the embodiments and features of the embodiments may be arbitrarily combined with each other without conflict.
In the field of jewelry jade identification, the spectrum detection technology mainly comprises infrared spectrum analysis and laser Raman spectrum analysis. Since the inorganic material has strong centrosymmetric vibration mode, the infrared spectrum is not sensitive, while the Raman spectrum has obvious advantages. Raman spectrum is generally clearer, overlapping bands are rarely seen, and spectrogram analysis is more convenient. In addition, Raman spectroscopy has the advantages of being fast and accurate, generally not damaging the sample (solid, semi-solid, liquid or gas) during measurement, and simple or even unnecessary for sample preparation. Therefore, the Raman spectrum is mainly used for research of inorganic minerals and nondestructive identification of gemstones and jades. By using the Raman spectrometer, the information such as components, crystal structures, molecular ligand structures and the like in the gem can be rapidly acquired, so that the variety and the authenticity of the gem can be identified; in addition, by analyzing the components of the gem inclusion, the origin of the gem and whether the gem is artificially processed or not can be judged; and by testing the photoluminescence spectrum (PL) of the gem, the crystal defects, the reasons thereof and the like of the gem can be analyzed. Compared with other technologies, the Raman spectrum is an effective, convenient, rapid and accurate analysis means for jewelry and jade products needing nondestructive testing, and the development space is huge. The laser micro-Raman spectrometer is used in the field of jewelry identification, has high identification accuracy, but has expensive equipment, complex operation and inconvenient carrying, and can only detect the jewelry jade species in a laboratory. Therefore, the handheld Raman spectrometer is widely applied to an identification field due to the characteristics of low manufacturing cost, convenience in use and the like, but is limited by equipment, the resolution of the collected Raman spectrum is not very high, meanwhile, due to the uncertainty of the sampling environment, the adopted personnel and the sampling specification, the handheld Raman spectrometer can hardly collect the Raman spectrum with high resolution, the gem can generate background fluorescence with different degrees under the excitation of laser, the interference of the fluorescence has great influence on the weak Raman scattering signals, the final Raman spectrum can not be imaged, and the accuracy of the detection result is influenced.
The natural gem jade has a variety of kinds, and can be divided into natural gem jade and artificial gem according to the gem composition, and different gems in the natural gem jade may have similar chemical compositions or chemical structures, thereby having similar Raman spectrograms, such as quartz jade and crystal. The traditional technology identifies the object to be detected according to the Raman spectrum, although the accurate identification can be carried out on most kinds of jewelry jades, a lot of uncertainty exists in the identification of a few jewelry jades with similar Raman spectrograms, the requirement on the algorithm is high when the similar Raman spectrums are to be accurately distinguished, the accurate distinguishing is difficult to carry out based on a correlation coefficient algorithm and a common machine learning algorithm, the misjudgment is often generated, therefore, a set of algorithm related to the characteristics of the jewelry jades is required to be provided, the object to be detected can be accurately identified, and the accuracy of the identification is improved.
In order to solve the problems in the prior art, the jewelry jade identification method based on deep learning is provided. Prior to carrying out the identification method herein, known jewelry jades are classified, and existing jewelry jades are classified into a plurality of species categories, such as: mammoth teeth, ivory, jade, crystal, pearl, shell, etc. On the basis, the jewelry jades of a plurality of different species with similar Raman spectra are divided into a group, and the jewelry jades of the plurality of different species in the group are taken as the same similar species in the text. The specific grouping method and the grouping number are determined according to practical requirements and are not limited in the text.
FIG. 1 is a flow diagram illustrating a method for deep learning-based jewelry jade qualification according to an exemplary embodiment. Referring to fig. 1, the jewelry jade identification method based on deep learning is applied to a raman spectrometer and comprises the following steps:
step S11, a raman spectrum of the analyte is acquired.
And step S12, inputting the Raman spectrum into a first-level identification model, and determining the similar category or the species category of the object to be detected.
And step S13, if the object to be detected is of a similar type, selecting a second-stage identification model corresponding to the similar type, inputting the Raman spectrum into the corresponding second-stage identification model, and outputting the species type of the object to be detected.
And step S11, acquiring the Raman spectrum of the object to be detected by the Raman spectrometer for the object to be detected. In this embodiment, the raman spectrum of the substance to be measured may be obtained by the laser micro-raman spectrometer, or may be obtained by the handheld raman spectrometer.
And step S12, inputting the Raman spectrum into the first-stage identification model, and determining the similar category or the species category of the object to be detected.
The jewel jades are peculiar in that most of the jewel jades are aggregates (in common words, mixtures), and for some jewels, the content of chemical components of different samples in the same species can be changed, that is, the spectrum of the jewel jades in the same species can be greatly changed, for example, the jade jades with various colors have the same chemical components, but the color is different because the proportion of other trace mineral components in the aggregates is different, and the corresponding Raman spectrum is also different. There are also gemstones, although of different species, with similar chemical compositions or structures, such as the silica species gemstones, whose raman spectra are very close due to the similarity of their main chemical compositions. For another example, the rosasite is associated with quartz, the mineral composition is complex, and different parts of the same sample have different raman spectra. When the traditional Raman spectrometer identifies the substances, an accurate identification conclusion cannot be given, and professional technicians are needed to comprehensively analyze the samples in the later period, so that the efficiency is low, and the accuracy is not high.
Since the raman spectrum of the analyte is related to the chemical composition and chemical structure of the analyte, in this embodiment, a group of substances with similar chemical compositions or chemical structures are classified into a similar category. For example, mammoth ivory and ivory are different species classes but have similar chemical compositions, and herein they are divided into a group, and the substances of a plurality of species classes in a group are taken as a similar class, which is also called mammoth ivory-ivory; for another example, the multiflower rose pyroxene and quartz are classified into a similar category, which is also called multiflower pyroxene-quartz jade; the agate is a mineral mixed with cryptocrystalline quartz, while the crystal is a quartz crystal with similar chemical structure, and the agate and the crystal are divided into similar categories, which are also called as quartz jade-crystal categories.
In the identification process, for most of jewelry jades with unique spectral characteristics, specific species can be identified through a first-stage identification model; for substances with similar Raman spectrograms, the species class cannot be directly judged, and the similar class of the object to be detected needs to be identified first.
On the basis, in step S13, if the object to be measured is of a certain similar type, the second-stage identification model corresponding to the similar type is selected, the raman spectrum is input into the corresponding second-stage identification model, and the species type of the object to be measured is output.
If the object to be detected is of a certain similar type, the object to be detected and other substances have similar Raman spectra, and accurate prediction cannot be carried out through the first-stage identification model. For example, in the recognition of the target dental product, only the dental product is recognized as a mammoth ivory-ivory based on the first-stage recognition model and the raman spectrum, and it is not possible to determine whether the dental product is a mammoth ivory or an ivory. Therefore, the object to be detected with similar Raman spectra is further identified through the second-stage identification model, so that different jewelry jades with similar Raman spectra are accurately distinguished. For example, when the raman spectrum of the dental product of the subject is identified, the first-stage identification model identifies the raman spectrum, and the features of the raman spectrum conform to the features of the mammoth-ivory, so that the object to be detected is identified to be of a mammoth-ivory similar type, and the species category may be the mammoth and the ivory. And then selecting a second-stage identification model corresponding to the similar species of the mammoth-ivory, and determining the species of the mammoth-ivory or the ivory by the second-stage identification model to realize accurate identification.
In an exemplary embodiment, the first-level identification model is a model obtained through deep learning method training, and is trained by using sample data of the jewelry jades marked with similar categories or species categories, and is used for identifying whether the object to be detected is the jewelry jades or not and determining the similar categories or species categories of the object to be detected. As most of the jewelry jades are mixtures, similar categories or species categories of the objects to be detected are identified through a first-level identification model. For example, raman spectra of a plurality of known articles are collected as training samples and the samples are labeled. For example, there are no materials with similar raman spectra, labeled with the species class of the material itself; materials with similar raman spectra, such as mammoth ivory and ivory products, the raman spectra of mammoth ivory products, the raman spectra of ivory products are labeled mammoth ivory-ivory class; similarly, the Raman spectrum of the agate product and the Raman spectrum of the crystal product are marked as the quartzite-crystal class; the name of the category may be a numeric identifier or a text identifier, and each similar category needs to be uniquely identified, and the specific identification method may be determined according to a specific use environment and use, which is not limited herein. Of course, the classification of substance classes shown herein is merely illustrative, and in practical applications, classification of the jewelry jade is not limited to the above.
And training the first-stage recognition model by using the marked samples. For example, the sample of the present embodiment includes a plurality of substances having similar raman spectra, and further includes a plurality of substances not having similar raman spectra. For a plurality of substances with similar Raman spectra, grouping the substances according to the similarity degree of the Raman spectra, wherein the Raman spectra of the substances in the same group are marked as the same similar category; for substances that do not have a similar raman spectrum, each substance is labeled with the species of the substance. And (4) performing iterative training on the first-stage recognition model for more than 100 times when the number of the sampling samples is more than 300 until the first-stage recognition model reaches a preset recognition accuracy rate.
In an exemplary embodiment, the second-stage recognition model comprises a plurality of models obtained after deep learning or other machine learning methods, and each model is trained by using sample data of a group of jewelry jades with similar raman spectra and labeled with a species class for recognizing the species class of the object to be detected. And selecting substances of the same similar type, similar chemical components or similar chemical structures and different species for sampling, acquiring enough training samples with similar Raman spectra, and training the deep learning model by using the training samples until the preset identification accuracy is reached. Respectively sampling a plurality of similar categories of substances, training a deep learning model, and establishing an invisible corresponding relation between the trained machine learning model and the similar categories of the training samples so as to obtain a plurality of second-stage identification models, wherein the number of the second-stage identification models is determined by the group number of the similar categories defined in the grouping method and is in one-to-one correspondence with the divided similar categories.
Based on the particularity of the jewelry jade, after the similar category of the object to be detected is determined, the object to be detected with the similar category can be determined to be the specific species category in the similar category through the second-stage identification model. If the first-stage identification model identification result is quartz jade-crystal during the identification of a certain mineral substance, selecting a second-stage identification model corresponding to the quartz jade-crystal, inputting the Raman spectrum into the selected second-stage identification model again, and identifying the object to be detected as crystal or quartz jade by the second-stage identification model. And for another example, identifying the ornament, namely, identifying the ornament as the mammoth ivory or the mammoth ivory by selecting a second-stage identification model corresponding to the mammoth ivory as a first-stage identification model identification result.
The jewelry jade identification method based on deep learning can be applied to a handheld Raman spectrometer, can still achieve high identification accuracy rate under the condition that the resolution of the collected Raman spectrum is not very high, and overcomes the defects of the traditional handheld Raman spectrometer.
For example, the following steps are carried out:
for organic gemstones, the ivory and mammoth ivory are very close in chemical composition and therefore also very close in raman spectrogram. FIG. 2 is a schematic representation of a Raman spectrum of an ivory tooth shown in accordance with an exemplary embodiment; fig. 3 is a schematic raman spectrum of a mammoth ivory shown in accordance with an exemplary embodiment. Comparing fig. 2 and fig. 3, it can be seen that it is difficult to distinguish the two substances by comparatively analyzing raman spectra of the two substances. Therefore, the traditional Raman spectrum identification algorithm is difficult to distinguish the two substances, and the jewelry jade identification method based on deep learning provided by the text can accurately identify the two substances through a machine learning model constructed by a multi-stage artificial intelligence algorithm, and the accuracy can reach more than 90%.
For the silicon dioxide jewelry jade variety, the main chemical components are similar, so that the distinction is difficult. Fig. 4 is a schematic diagram of raman spectra of agate and crystal, shown in accordance with an example embodiment. The raman spectra of agate and crystal shown in fig. 4 are very similar, because agate is a mineral mixed with cryptocrystalline quartz, and crystal is a quartz crystal, so that they have very similar main raman spectra and are located near 460cm-1, and if such similar spectra are to be accurately distinguished, the algorithm requirement is very high, and accurate distinguishing is difficult to be achieved based on a correlation coefficient algorithm and a common non-deep learning machine learning algorithm, and misjudgment can be generated. The jewelry jade identification method based on deep learning can greatly improve the discrimination capability, and the identification accuracy can reach 97%.
Aiming at the specific characteristics of the jewelry jade, a hierarchical identification method is adopted for identifying the jewelry jade, the similar category or the species category of the object to be detected is determined by a first-stage identification model, the substance identified as the similar category is selected by a corresponding second identification model, and the Raman spectrum is identified for the second time to identify the specific species category of the object to be detected. The method provided by the invention can quickly identify the species of most substances, and can effectively improve the identification accuracy for substances of a plurality of different species with similar Raman spectra.
Fig. 5 is a block diagram illustrating a deep learning based jewelry jade appraisal apparatus according to an exemplary embodiment. Referring to fig. 5, the jewelry jade appraisal apparatus based on deep learning includes: a raman spectrum acquisition module 501, a first identification module 502, a model selection module 503, and a second identification module 504.
The raman spectrum acquiring module 501 is configured to acquire a raman spectrum of an object to be measured.
The first identification module 502 is configured to input the raman spectrum into a first-stage identification model, and determine a similar class or species class of the analyte.
The model selection module 503 is configured to select a second-level recognition model corresponding to a similar category if the object to be tested is the similar category.
The second identification module 504 is configured to input the raman spectrum into the corresponding second-stage identification model, and output the species to be detected as the genus class.
The first identification module is used for training the model obtained after deep learning method training by using the sample data of the jewelry jades marked with similar categories or species categories, identifying whether the object to be detected is the jewelry jades or not, and determining the similar categories or species categories of the object to be detected.
The second identification module comprises a plurality of models obtained after deep learning or other machine learning methods are trained, and each model is trained by using sample data of a group of jewelry jades with similar Raman spectra marked with species classes and is used for identifying the species classes of the object to be detected.
As will be appreciated by one skilled in the art, the embodiments herein may be provided as a method, apparatus (device), or computer program product. Accordingly, the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied in the medium. Computer storage media includes 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, including, but not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, Digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by the computer, and the like. In addition, communication media typically embodies 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 as known to those skilled in the art.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (devices) and computer program products according to embodiments herein. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that an article or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such article or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of additional like elements in the article or device comprising the element.
While the preferred embodiments herein have been described, additional variations and modifications of these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following appended claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of this disclosure.
It will be apparent to those skilled in the art that various changes and modifications may be made herein without departing from the spirit and scope thereof. Thus, it is intended that such changes and modifications be included herein, provided they come within the scope of the appended claims and their equivalents.
Claims (10)
1. A jewelry jade identification method based on deep learning is applied to a Raman spectrometer and is characterized by comprising the following steps:
acquiring a Raman spectrum of an object to be detected;
inputting the Raman spectrum into a first-stage identification model, and determining the similar category or the species category of the object to be detected;
and if the object to be detected is of a certain similar type, selecting a second-stage identification model corresponding to the similar type, inputting the Raman spectrum into the corresponding second-stage identification model, and outputting the species type of the object to be detected.
2. The deep learning-based jewelry jade identification method of claim 1, further comprising: and classifying the jewelry jades of different species with similar Raman spectra into the same similar species.
3. The deep learning-based jewelry jade identification method of claim 1, wherein said raman spectrometer comprises: a hand-held Raman spectrometer.
4. The method for identifying jewelry jade based on deep learning of claim 1, wherein the first-stage recognition model is a model obtained after deep learning method training, and is trained by using sample data of jewelry jade marked with similar categories or species categories for recognizing whether the object to be detected is jewelry jade and determining the similar categories or species categories of the object to be detected.
5. The method for identifying jewelry jades based on deep learning of claim 1, wherein the second-stage recognition model comprises a plurality of models obtained after deep learning or other machine learning methods, and each model is trained by using sample data of a group of jewelry jades with similar Raman spectra marked with species classes for recognizing the species classes of the object to be tested.
6. A jewelry jade appraisal device based on deep learning, comprising:
the Raman spectrum acquisition module is used for acquiring a Raman spectrum of the object to be detected;
the first identification module is used for inputting the Raman spectrum into a first-stage identification model and determining the similar category or the species category of the object to be detected;
the model selection module is used for selecting a second-level identification model corresponding to a similar category if the object to be detected is of the similar category;
and the second identification module is used for inputting the Raman spectrum into the corresponding second-stage identification model and outputting the species to be detected.
7. The jewelry jade appraising device based on deep learning of claim 6, wherein the first identification module is trained by using sample data of jewelry jade marked with similar categories or species categories for a model obtained after deep learning method training, and is used for identifying whether the object to be detected is jewelry jade and determining the similar categories or species categories of the object to be detected.
8. The deep learning-based jewelry jade appraisal device of claim 6, wherein the second identification module comprises a plurality of models obtained after deep learning or other machine learning methods training, each model being trained using sample data of a group of jewelry jades with similar Raman spectra marked with a species class for identifying the species class of the object to be tested.
9. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed, implements the steps of the method according to any one of claims 1-5.
10. A computer arrangement comprising a processor, a memory and a computer program stored on the memory, characterized in that the steps of the method according to any of claims 1-5 are implemented when the computer program is executed by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110240147.8A CN113008865A (en) | 2021-03-04 | 2021-03-04 | Method, device, medium and equipment for identifying jewelry jade |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110240147.8A CN113008865A (en) | 2021-03-04 | 2021-03-04 | Method, device, medium and equipment for identifying jewelry jade |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113008865A true CN113008865A (en) | 2021-06-22 |
Family
ID=76405241
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110240147.8A Pending CN113008865A (en) | 2021-03-04 | 2021-03-04 | Method, device, medium and equipment for identifying jewelry jade |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113008865A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113640272A (en) * | 2021-07-13 | 2021-11-12 | 黄埔海关技术中心 | Spectral library establishing and detecting method based on portable Raman spectrometer and application |
| CN115905881A (en) * | 2022-09-30 | 2023-04-04 | 国家珠宝玉石首饰检验集团有限公司 | Method and device for classifying yellow pearls, electronic equipment and storage medium |
| CN116818761A (en) * | 2023-08-16 | 2023-09-29 | 国检中心深圳珠宝检验实验室有限公司 | Method and equipment for detecting color cause of single crystal color precious stone |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103837518A (en) * | 2012-11-20 | 2014-06-04 | 福州高意通讯有限公司 | Raman substance recognition platform based on cloud computing |
| CN107037028A (en) * | 2017-03-10 | 2017-08-11 | 北京华泰诺安探测技术有限公司 | A kind of cloud platform Raman spectrum recognition methods and device |
| WO2019127352A1 (en) * | 2017-12-29 | 2019-07-04 | 深圳达闼科技控股有限公司 | Raman spectrum-based substance identification method and cloud system |
| CN112149754A (en) * | 2020-09-30 | 2020-12-29 | 罗中岩 | Information classification method, device, equipment and storage medium |
-
2021
- 2021-03-04 CN CN202110240147.8A patent/CN113008865A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103837518A (en) * | 2012-11-20 | 2014-06-04 | 福州高意通讯有限公司 | Raman substance recognition platform based on cloud computing |
| CN107037028A (en) * | 2017-03-10 | 2017-08-11 | 北京华泰诺安探测技术有限公司 | A kind of cloud platform Raman spectrum recognition methods and device |
| WO2019127352A1 (en) * | 2017-12-29 | 2019-07-04 | 深圳达闼科技控股有限公司 | Raman spectrum-based substance identification method and cloud system |
| CN112149754A (en) * | 2020-09-30 | 2020-12-29 | 罗中岩 | Information classification method, device, equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 汪舒: "激光拉曼光谱分析在珠宝鉴定中的应用进展", 《中小企业管理与科技(中旬刊)》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113640272A (en) * | 2021-07-13 | 2021-11-12 | 黄埔海关技术中心 | Spectral library establishing and detecting method based on portable Raman spectrometer and application |
| CN115905881A (en) * | 2022-09-30 | 2023-04-04 | 国家珠宝玉石首饰检验集团有限公司 | Method and device for classifying yellow pearls, electronic equipment and storage medium |
| CN115905881B (en) * | 2022-09-30 | 2023-10-20 | 国家珠宝玉石首饰检验集团有限公司 | Yellow pearl classification method and device, electronic equipment and storage medium |
| CN116818761A (en) * | 2023-08-16 | 2023-09-29 | 国检中心深圳珠宝检验实验室有限公司 | Method and equipment for detecting color cause of single crystal color precious stone |
| CN116818761B (en) * | 2023-08-16 | 2024-04-19 | 国检中心深圳珠宝检验实验室有限公司 | Method and equipment for detecting color cause of single crystal color precious stone |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113008865A (en) | Method, device, medium and equipment for identifying jewelry jade | |
| JP6091493B2 (en) | Spectroscopic apparatus and spectroscopy for determining the components present in a sample | |
| CN104807787B (en) | A kind of Classification of Tea discrimination method based on LIBS | |
| CN103439266A (en) | Identification and analysis system for jewelry jade and rock minerals | |
| US10718713B2 (en) | Unknown sample determining method, unknown sample determining instrument, and unknown sample determining program | |
| WO2018121082A1 (en) | Self-learning-type qualitative analysis method based on raman spectrum | |
| CN108780048A (en) | A kind of method, detection device and the readable storage medium storing program for executing of determining detection device | |
| CN105717066A (en) | Near-infrared spectrum recognition model based on weighting association coefficients | |
| CN104345045A (en) | Chemical pattern recognition and near infrared spectrum-based similar medicinal material identification method | |
| CN111259929A (en) | Random forest based food-borne pathogenic bacteria classification model training method | |
| CN115618282A (en) | Synthetic gem identification method, device and storage medium | |
| CN105223140A (en) | The method for quickly identifying of homology material | |
| CN111426657B (en) | Identification comparison method of three-dimensional fluorescence spectrogram of soluble organic matter | |
| CN108169204B (en) | Raman spectrum preprocessing method based on database | |
| US20150062575A1 (en) | Method for measuring performance of a spectroscopy system | |
| CN113125374B (en) | Method, device and equipment for detecting REE content in carbonate type rare earth deposit sample | |
| CN113791062A (en) | Method for judging fixed substance type based on Raman spectrum | |
| CN115078327A (en) | Rapid detection method for dangerous chemicals based on neural network architecture search | |
| AU2019100481A4 (en) | Quick and nondestructive cubilose authenticity discrimination method | |
| Lowry et al. | The use of a Raman spectral database of minerals for the rapid verification of semiprecious gemstones | |
| CN106404696A (en) | Spectrum detection method for performing species identification on honey | |
| CN109145887B (en) | Threshold analysis method based on spectral latent variable confusion discrimination | |
| Peltola | IT-time for mechanized forest work study | |
| CN113640277B (en) | Method for rapidly identifying eutectic explosive structure based on chemometrics | |
| CN113030060A (en) | Drug Raman spectrum identification method based on convolutional neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210622 |