CN114239758B - Jewelry money code determination method, jewelry money code determination device, computer equipment and storage medium - Google Patents
Jewelry money code determination method, jewelry money code determination device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN114239758B CN114239758B CN202210174482.7A CN202210174482A CN114239758B CN 114239758 B CN114239758 B CN 114239758B CN 202210174482 A CN202210174482 A CN 202210174482A CN 114239758 B CN114239758 B CN 114239758B
- Authority
- CN
- China
- Prior art keywords
- jewelry
- identification
- sample
- picture
- mapping relation
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The application relates to a jewelry code determination method, a jewelry code determination device, a computer device and a storage medium. The method comprises the following steps: acquiring a jewelry picture; the jewelry picture comprises jewelry to be identified; obtaining jewelry feature vectors according to the jewelry pictures; determining a jewelry identifier corresponding to the jewelry to be recognized through the jewelry feature vector according to a preset first mapping relation; the first mapping relation is a mapping relation between the jewelry identification and the jewelry characteristic vector; determining a jewelry money code corresponding to the jewelry to be identified through the jewelry identification according to a preset second mapping relation; the second mapping relation is the mapping relation between the jewelry identification and the jewelry money code. By adopting the method, the corresponding relation between a large number of jewelry characteristics and jewelry codes does not need to be memorized manually, and the daily use of the codes is convenient.
Description
Technical Field
The present application relates to the field of data processing technologies, and in particular, to a jewelry style code determining method, apparatus, computer device, and storage medium.
Background
The money code is a string of codes generated according to the characteristics of the jewelry, and generally consists of numbers and characters. In the jewelry selling and purchasing process, money codes are set for the jewelry, the jewelry characteristics can be described quickly and accurately, a jewelry selling store can conveniently search the inventory according to the requirements of customers, and a purchaser purchases the jewelry according to a specific style.
At present, in a money code application process, when a withdrawal code needs to be obtained according to a specified jewelry characteristic, money codes corresponding to various styles are generated by means of manual memory, for example, when a clerk in a jewelry sales store needs to search for the inventory of gold lily rings, money codes corresponding to gold lily, lily and the ring need to be memorized respectively, and then the money codes of the gold lily rings are generated in a combined manner. This requires a lot of manual memory, which is inconvenient for the routine use of the bar code. In addition, when a store clerk needs to identify jewelry characteristics manually and then generate a money code corresponding to a designated jewelry, the manual identification is easily influenced by subjective factors such as store clerk experience, so that the characteristic identification is easily inaccurate and the accuracy of the money code generation is further influenced.
Therefore, the current jewelry code acquisition technology has the problem of inconvenient daily use.
Disclosure of Invention
In view of the foregoing, there is a need to provide a jewelry code determination method, apparatus, computer device, and computer-readable storage medium that can facilitate daily use.
In a first aspect, the present application provides a jewelry code determination method. The method comprises the following steps:
obtaining a jewelry picture; the jewelry picture comprises jewelry to be identified;
obtaining jewelry feature vectors according to the jewelry pictures;
determining a jewelry identification corresponding to the jewelry to be recognized through the jewelry characteristic vector according to a preset first mapping relation; the first mapping relation is a mapping relation between the jewelry identification and the jewelry characteristic vector;
determining a jewelry money code corresponding to the jewelry to be identified through the jewelry identification according to a preset second mapping relation; the second mapping relation is the mapping relation between the jewelry identification and the jewelry money code.
In one embodiment, the jewelry picture is at least one; according to the jewelry picture, a jewelry feature vector is obtained, and the method comprises the following steps:
inputting each jewelry picture into a pre-trained feature extraction network to obtain at least one picture feature vector output by the feature extraction network;
and averaging the at least one picture feature vector to obtain the jewelry feature vector.
In one embodiment, the method further comprises:
obtaining at least one picture of a jewelry sample;
determining a sample feature vector according to at least one picture of the jewelry sample;
obtaining a sample identification of the jewelry sample;
and establishing a corresponding relation between the sample identification and the sample characteristic vector to obtain the first mapping relation.
In one embodiment, the method further comprises:
identifying sample features from at least one picture of the jewelry sample;
determining a sample money code of the jewelry sample according to the sample characteristics;
and establishing a corresponding relation between the sample identification and the sample money code to obtain the second mapping relation.
In one embodiment, the identifying sample characteristics from the at least one picture of the jewelry sample comprises:
inputting each picture into a pre-trained feature recognition model to obtain at least one first recognition result output by the feature recognition model; the first identification result comprises a first identification result of jewelry types, a first identification result of diamond types, a first identification result of metal materials and a first identification result of surface processes;
obtaining a second jewelry type identification result according to the first jewelry type identification result which is identified most;
obtaining a second identification result of the diamond types according to the first identification results of all the identified diamond types;
obtaining a second identification result of the metal material according to the first identification result of the most identified metal materials;
obtaining a second recognition result of the surface process according to the pixel number corresponding to each first recognition result of the surface process;
and obtaining the sample characteristics according to the second jewelry type identification result, the second diamond type identification result, the second metal material identification result and the second surface process identification result.
In one embodiment, the method further comprises:
obtaining a first jewelry identification of a first jewelry and a second jewelry identification of a second jewelry;
determining a first jewelry characteristic vector corresponding to the first jewelry identification and a second jewelry characteristic vector corresponding to the second jewelry identification according to the first mapping relation;
calculating a vector distance between the first jewelry feature vector and the second jewelry feature vector;
and if the vector distance is smaller than a preset threshold value, judging that the first jewelry and the second jewelry are in the same style.
In a second aspect, the present application also provides a jewelry item code determination apparatus. The device comprises:
the picture acquisition module is used for acquiring jewelry pictures; the jewelry picture comprises jewelry to be identified;
the characteristic vector determining module is used for obtaining a jewelry characteristic vector according to the jewelry picture;
the identification determining module is used for determining jewelry identification corresponding to the jewelry to be recognized through the jewelry characteristic vector according to a preset first mapping relation; the first mapping relation is a mapping relation between the jewelry identification and the jewelry characteristic vector;
the money code determining module is used for determining jewelry money codes corresponding to the jewelry to be recognized through the jewelry identification according to a preset second mapping relation; the second mapping relation is the mapping relation between the jewelry identification and the jewelry money code.
In a third aspect, the application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the following steps when executing the computer program:
acquiring a jewelry picture; the jewelry picture comprises jewelry to be identified;
obtaining jewelry feature vectors according to the jewelry pictures;
determining a jewelry identifier corresponding to the jewelry to be recognized through the jewelry feature vector according to a preset first mapping relation; the first mapping relation is a mapping relation between the jewelry identification and the jewelry characteristic vector;
determining a jewelry money code corresponding to the jewelry to be identified through the jewelry identification according to a preset second mapping relation; the second mapping relation is the mapping relation between the jewelry identification and the jewelry money code.
In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:
obtaining a jewelry picture; the jewelry picture comprises jewelry to be identified;
obtaining jewelry feature vectors according to the jewelry pictures;
determining a jewelry identifier corresponding to the jewelry to be recognized through the jewelry feature vector according to a preset first mapping relation; the first mapping relation is a mapping relation between the jewelry identification and the jewelry characteristic vector;
determining a jewelry money code corresponding to the jewelry to be recognized through the jewelry identification according to a preset second mapping relation; the second mapping relation is the mapping relation between the jewelry identification and the jewelry money code.
The jewelry money code determining method, the device, the computer equipment and the storage medium are characterized in that a jewelry picture is obtained, jewelry characteristic vectors are obtained according to the jewelry picture, the characteristics of jewelry to be recognized can be digitally described, jewelry identification corresponding to the jewelry to be recognized is determined through the jewelry characteristic vectors according to a preset first mapping relation, identification of the jewelry to be recognized can be conveniently and quickly obtained through the characteristic vectors, jewelry money codes corresponding to the jewelry to be recognized can be determined through the jewelry identification according to a preset second mapping relation, money codes of the jewelry to be recognized can be conveniently and quickly obtained through the identification, a large number of corresponding relations between jewelry characteristics and the jewelry money codes do not need to be manually memorized, and the daily use of the money codes is facilitated.
Moreover, the jewelry features can be digitally described by acquiring the jewelry picture of the jewelry to be identified, obtaining the jewelry feature vector according to the jewelry picture and determining the money code by using the feature vector, so that the accuracy of feature identification is ensured.
Drawings
FIG. 1 is a schematic flow chart of a jewelry item code determination method in one embodiment;
FIG. 2 is a schematic flow chart of a jewelry item code determination method in another embodiment;
FIG. 3 is a block diagram showing the construction of a jewelry code determining apparatus according to an embodiment;
FIG. 4 is a diagram of the internal structure of a computer device in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The jewelry style code determining method provided by the embodiment of the application can be applied to a terminal, the terminal can be but is not limited to various personal computers, notebook computers, smart phones, tablet computers, internet of things equipment and portable wearable equipment, and the internet of things equipment can be smart sound boxes, smart televisions, smart air conditioners, smart vehicle-mounted equipment and the like. The portable wearable device can be a smart watch, a smart bracelet, a head-mounted device, and the like.
In one embodiment, as shown in fig. 1, a jewelry code determining method is provided, which is described by taking an example of applying the method to a terminal, and includes the following steps:
step S110, acquiring a jewelry picture; the jewelry picture comprises jewelry to be identified.
In the concrete realization, can put the jewelry of treating discernment on the unified platform of background, under stable external light condition, treat discernment jewelry from fixed distance and shoot, can only shoot a jewelry picture, also can change the jewelry at the in-process of shooing and put the gesture, shoot many jewelry pictures from different angles. The number of the jewelry pictures shot can be recorded as M (M is more than or equal to 1).
The jewelry to be identified can be shot by using the camera equipment carried by the terminal, and the jewelry can also be shot by using the external camera equipment, and the shot jewelry picture is transmitted to the terminal through a network or a data line. In order to improve the shooting quality, the camera equipment and the external camera equipment of the terminal can be special camera equipment for jewelry shooting.
And step S120, obtaining jewelry feature vectors according to the jewelry pictures.
In specific implementation, the terminal can input each of the M jewelry pictures into a trained feature extraction network of a convolutional neural network structure, and for each jewelry picture, the feature extraction network can output a feature vector with a specified length (for example, the length is 512), so as to obtain M feature vectors, and average the M feature vectors, so as to obtain the jewelry feature vector of the jewelry to be recognized.
Step S130, according to a preset first mapping relation, determining a jewelry identification corresponding to the jewelry to be identified through a jewelry characteristic vector; the first mapping relationship is a mapping relationship between the jewelry identifications and the jewelry feature vectors.
In a specific implementation, a lookup table recorded with mapping relationships between jewelry identifications and jewelry feature vectors can be generated in advance and stored in a database of the terminal as a first mapping relationship. After the jewelry characteristic vector is obtained, the jewelry identification corresponding to the jewelry characteristic vector can be searched in the database according to the first mapping relation and used as the jewelry identification of the jewelry to be identified.
The jewelry identification in the first mapping relation can be randomly generated, jewelry characteristic vectors can be obtained by collecting a plurality of jewelry samples, collecting at least one picture of each jewelry sample, recording the number of the pictures as N (N is more than or equal to 1), inputting each picture into a trained characteristic extraction network with a convolutional neural network structure to obtain N characteristic vectors, and averaging the N characteristic vectors to obtain the characteristic vector of each jewelry sample. And recording the corresponding relation between the identification of each jewelry sample and the characteristic vector in a database to obtain a first mapping relation.
Step S140, determining jewelry money codes corresponding to the jewelry to be recognized through jewelry identification according to a preset second mapping relation; the second mapping relation is a mapping relation between the jewelry identification and the jewelry item code.
In a specific implementation, a lookup table recorded with the mapping relationship between the jewelry identification and the jewelry money code may be generated in advance, and stored in the database of the terminal as the second mapping relationship. After the jewelry identification is obtained, the jewelry money code corresponding to the jewelry identification can be searched in the database according to the first mapping relation and used as the money code of the jewelry to be identified.
The jewelry mark in the second mapping relation can be randomly generated, and the jewelry money code can be generated according to the jewelry type, whether a diamond and a diamond type are inlaid on the jewelry, the metal material of the jewelry, the surface process mode of the jewelry and other characteristics. The second mapping relationship may be obtained by obtaining the identifiers and the codes of the plurality of jewelry samples, and recording the correspondence between the identifier and the code of each jewelry sample in a database.
In practical application, the jewelry sample can be extracted with the following characteristics: (1) jewelry varieties including ring, pendant, bracelet, necklace and the like; (2) whether the diamond and the diamond type are included, wherein the diamond type comprises inlaying modes such as claw inlaying, wrapping inlaying and wall inlaying; (3) the metal material comprises two materials of gold and silver; (4) the surface process mode comprises surface processes such as smoothing, wire drawing, sanding, sand pushing and the like.
The jewelry sample is placed on a platform with a uniform background, a special camera device is used, the jewelry is photographed at a fixed distance under the condition of stable external light, the posture of the jewelry is changed in the photographing process so as to obtain a plurality of images photographed from different angles, and the number of the photographed images is recorded as N (N is more than or equal to 1).
Aiming at different jewelry characteristics, the primary identification is carried out on the shot jewelry sample picture by using different identification models respectively, and the method specifically comprises the following steps:
(a) preliminarily identifying the jewelry types: and recording the number of the jewelry types as K1, and inputting each of the N jewelry sample pictures into a classification network with a trained convolutional neural network structure. For each picture, the classification network may output a vector of length K1, where the sum of K1 numbers in the vector is 1, representing the probability of the jewelry sample picture being divided into each category, respectively. And taking the jewelry category with the highest probability as a primary identification result of the jewelry category.
(b) Whether diamond and diamond type primary identification are included: the number of the types of the diamond inlaying modes is K2, each of N jewelry sample pictures is input into a detection network with a trained convolutional neural network structure, the detection network outputs a plurality of detection frames, each detection frame comprises four coordinates representing the position of the detection frame and K2 numbers representing the probability of the inlaying types, and the number of K2 numbers is added to be 1. And for each output detection frame, taking the mosaic type with the highest probability as a preliminary identification result of the mosaic type.
(c) Metal material preliminary identification: the number of the types of the metal materials is K3, and each of the N jewelry sample pictures is input into a classification network with a trained convolutional neural network structure. For each picture, the classification network outputs a vector with a length of K3, wherein the sum of K3 numbers in the vector is 1, which respectively represents the probability that the jewelry sample picture is classified into each metal material. And taking the metal material with the highest probability as a primary identification result of the metal material.
(d) Preliminarily identifying a surface process mode: the number of types of surface process modes is K4, and each of the N jewelry sample pictures is input into a semantic segmentation network with a trained convolutional neural network structure. For each picture, the semantic segmentation network outputs a feature map with the same size as the input picture, each pixel point on the feature map is a vector with the length of K4, and the sum of K4 numbers in the vector is 1, which respectively represents the probability that the pixel point is divided into each surface process mode. And taking the surface process mode with the highest probability as the surface process mode of the pixel point, counting the number of the pixel points respectively belonging to each surface process mode, and taking the counted number as a primary identification result of the surface process mode.
In the above recognition model, the convolutional neural network is a machine learning model, and can be obtained by a data training method, and the specific model training process may include:
step S101, collecting data. In the environment as close as possible to the actual use scene, a large number of pictures are taken of various jewelry in the same shooting mode as in actual use. It should be noted that the categories and styles of jewelry photographed should cover as much jewelry as possible photographed in actual use scenarios.
And S102, marking data. Manually marking on the shot picture, wherein the marking modes of each identification task can be as follows: (1) extracting jewelry feature vectors: marking pictures of the same jewelry as a category; (2) the jewelry category: marking each picture with a jewelry category respectively; (3) whether diamond is involved and the type of diamond: on each picture, each diamond is framed in a rectangular frame mode, and a diamond embedding mode is marked for each frame; (4) the metal material is as follows: respectively marking a metal material type for each picture; (5) the surface process mode is as follows: on each picture, a polygonal frame mode is adopted respectively, areas of the same surface process, which are connected together, are framed, and the surface process type is marked for each polygonal frame.
And step S103, training a model. And (4) randomly dividing the data set labeled in the step S102 into 90% and 10% data sets, wherein 90% of the data sets are used for model training, and 10% of the data sets are used for model testing. And respectively training the convolutional neural network models of the recognition tasks based on the training data.
And step S104, parameter adjustment and model selection. And adjusting model parameters, respectively training different recognition models, comparing the effect of each recognition model on the test data set, and selecting the model with the highest recognition accuracy on the test data set, wherein the model is the model in the final actual use scene.
After obtaining the preliminary recognition result, can carry out the fusion of recognition result based on the preliminary recognition result, obtain the final recognition result of jewelry characteristic, specifically include:
(e) the jewelry category: n jewelry sample pictures have exported N preliminary recognition results to this jewelry sample kind, and the number of the preliminary recognition result that each jewelry kind corresponds is counted, gets the kind that preliminary recognition result number is the most, as the final recognition result of jewelry sample kind.
(f) Whether diamond is involved and diamond type: and taking all appeared mosaic modes in the primary detection results of the N jewelry sample pictures as the diamond types of the jewelry samples. If no diamond detection frame is output in the N pictures, the jewelry sample is considered to have no diamond.
(g) The metal material is as follows: n jewelry sample pictures have exported to this jewelry sample metal material's a N preliminary recognition result, and the metal material that gets the most metal material of preliminary recognition result number is counted to the preliminary recognition result number that each metal material corresponds, as jewelry sample metal material's final recognition result.
(h) The surface process mode comprises the following steps: and respectively outputting the pixel number of each surface process mode by the N jewelry sample pictures, counting the pixel number sum corresponding to each surface process mode, comparing the obtained pixel number sum with a preset threshold value, and taking the surface process mode corresponding to the pixel number sum as the final identification result of the surface process mode if the obtained pixel number sum exceeds the preset threshold value.
The code of the jewelry sample can be generated according to the identified jewelry type, whether the diamond and the diamond type are inlaid on the jewelry, the metal material of the jewelry and the surface process mode of the jewelry. For example, a ring corresponding to the code 0001, a diamond-inlaid claw corresponding to the code 0010, a gold jewelry corresponding to the code 0100, and a plain surface corresponding to the code 1000 may be preset, and after the final recognition result that the jewelry type is the ring, the diamond-inlaid claw is inlaid on the jewelry, the metal material is gold, and the surface processing method is the plain surface is obtained, the code for generating the jewelry sample is 0001 + 0010 + 0100 + 1000.
The jewelry money code determining method comprises the steps of obtaining a jewelry picture, obtaining jewelry characteristic vectors according to the jewelry picture, digitally describing characteristics of jewelry to be recognized, determining jewelry marks corresponding to the jewelry to be recognized through the jewelry characteristic vectors according to a preset first mapping relation, conveniently and quickly obtaining the marks of the jewelry to be recognized through the characteristic vectors, determining the jewelry money codes corresponding to the jewelry to be recognized through the jewelry marks according to a preset second mapping relation, conveniently and quickly obtaining the money codes of the jewelry to be recognized through the marks, needing not to manually memorize corresponding relations between a large number of jewelry characteristics and the jewelry money codes, and facilitating daily use of the money codes.
Moreover, the jewelry features can be digitally described by acquiring the jewelry picture of the jewelry to be identified, obtaining the jewelry feature vector according to the jewelry picture and determining the money code by using the feature vector, so that the accuracy of feature identification is ensured.
In an embodiment, the step S120 may specifically include: inputting each jewelry picture into a pre-trained feature extraction network to obtain at least one picture feature vector output by the feature extraction network; and averaging at least one picture feature vector to obtain a jewelry feature vector.
The feature extraction network may be a convolutional neural network that performs feature vector extraction on the picture.
In specific implementation, each of the M jewelry pictures may be input into a trained feature extraction network having a convolutional neural network structure, and for each jewelry picture, the feature extraction network may output a feature vector of a specified length (for example, the length is 512), so as to obtain M feature vectors, and average the M feature vectors, so as to obtain jewelry feature vectors of jewelry to be recognized.
In this embodiment, each jewelry picture is input to the pre-trained feature extraction network to obtain at least one picture feature vector output by the feature extraction network, so that the efficiency of obtaining the picture feature vectors can be improved, the at least one picture feature vector is averaged to obtain the jewelry feature vectors, the feature vectors of the jewelry pictures shot from different angles can be fused, and the reliability of obtaining the feature vectors can be improved.
In an embodiment, the jewelry item code determining method may further include: obtaining at least one picture of a jewelry sample; determining a sample feature vector according to at least one picture of a jewelry sample; obtaining a sample identifier of a jewelry sample; and establishing a corresponding relation between the sample identification and the sample characteristic vector to obtain a first mapping relation.
In the specific implementation, a plurality of jewelry samples can be collected, at least one picture of each jewelry sample is collected, the number of the pictures is recorded as N (N is more than or equal to 1), each picture is input into a trained feature extraction network with a convolutional neural network structure to obtain N feature vectors, and the N feature vectors are averaged to obtain the feature vector of each jewelry sample, namely the sample feature vector. And randomly generating a sample identifier for each jewelry sample, and recording the corresponding relation between the sample identifier of each jewelry sample and the sample characteristic vector in a database to obtain a first mapping relation.
In this embodiment, by obtaining at least one picture of a jewelry sample, determining a sample feature vector according to the at least one picture of the jewelry sample, obtaining a sample identifier of the jewelry sample, and establishing a corresponding relationship between the sample identifier and the sample feature vector to obtain a first mapping relationship, the corresponding relationship between the jewelry sample identifier and the feature vector can be stored in a database, so that the sample identifier can be conveniently searched according to the feature vector, and the jewelry management efficiency is improved.
In an embodiment, the jewelry item code determining method may further include:
step S105, identifying sample characteristics according to at least one picture of the jewelry sample;
step S106, determining a sample money code of the jewelry sample according to the sample characteristics;
and step S107, establishing a corresponding relation between the sample identification and the sample money code to obtain a second mapping relation.
In specific implementation, each jewelry sample picture can be input into a trained recognition model with a convolutional neural network structure to obtain sample characteristics in the aspects of the jewelry sample type, whether a diamond is embedded, the diamond type, the metal material, the surface process mode and the like, and the codes of the jewelry sample are generated according to the codes corresponding to the sample characteristics, for example, the code corresponding to a ring 0001, the code corresponding to a claw-inlaid diamond 0010, the code corresponding to a gold jewelry 0100 and the code corresponding to a polished surface 1000 can be preset, when the jewelry type is recognized as a ring, the claw-inlaid diamond is inlaid on the jewelry, the metal material is gold, and after the surface process mode is a polished surface, the code of the jewelry sample can be generated as 0001-0010-0100-1000. And corresponding the identifier of each jewelry sample with the money code, and recording the identifier of each jewelry sample in a database to obtain a second mapping relation, wherein the identifier of each jewelry sample is generated randomly.
In this embodiment, the sample characteristics are identified according to at least one picture of the jewelry sample, the sample money code of the jewelry sample is determined according to the sample characteristics, the corresponding relationship between the jewelry sample identification and the sample money code is established by establishing the corresponding relationship between the sample identification and the sample money code, so that the second mapping relationship is obtained, the corresponding relationship between the jewelry sample identification and the money code can be stored in the database, the money code can be conveniently searched according to the sample identification, and the jewelry money code acquisition efficiency is improved.
In an embodiment, the step S105 may specifically include: inputting each picture into a pre-trained feature recognition model to obtain at least one first recognition result output by the feature recognition model; the first identification result comprises a first identification result of jewelry types, a first identification result of diamond types, a first identification result of metal materials and a first identification result of surface processes; obtaining a second jewelry type identification result according to the first jewelry type identification result which is identified most; obtaining a second identification result of the diamond types according to the first identification results of all the identified diamond types; obtaining a second metal material identification result according to the first identified metal material identification result with the most number; obtaining a second recognition result of the surface process according to the pixel number corresponding to the first recognition result of each surface process; and obtaining the sample characteristics according to the second identification result of the jewelry type, the second identification result of the diamond type, the second identification result of the metal material and the second identification result of the surface process.
In the concrete implementation, to different jewelry characteristics, can use different recognition models to carry out preliminary recognition to the jewelry sample picture of shooing respectively, obtain first recognition result, specifically include:
(a) preliminarily identifying the jewelry types: noting the number of jewelry types as K1, inputting each of the N jewelry sample pictures into a classification network with a trained convolutional neural network structure. For each picture, the classification network may output a vector of length K1, where the sum of K1 numbers in the vector is 1, representing the probability of the jewelry sample picture being divided into each category, respectively. And taking the jewelry category with the highest probability as a first identification result of the jewelry category.
(b) Whether diamond and diamond type primary identification are included: the number of types of diamond inlaying modes is K2, each of N jewelry sample pictures is input into a detection network with a trained convolutional neural network structure, the detection network outputs a plurality of detection frames, each detection frame comprises four coordinates representing the position of the detection frame and K2 numbers representing the probability of inlaying types, and the number of K2 numbers is added to be 1. And for each output detection frame, taking the mosaic type with the highest probability as a first identification result of the diamond type.
(c) Preliminarily identifying the metal material: and (4) recording the type number of the metal materials as K3, and inputting each of the N jewelry sample pictures into a classification network with a trained convolutional neural network structure. For each picture, the classification network outputs a vector with the length of K3, wherein the sum of K3 numbers in the vector is 1, and respectively represents the probability that the jewelry sample picture is divided into each metal material. And taking the metal material with the highest probability as a first identification result of the metal material.
(d) Preliminary identification of a surface process mode: the number of types of surface process modes is K4, and each of the N jewelry sample pictures is input into a semantic segmentation network with a trained convolutional neural network structure. For each picture, the semantic segmentation network outputs a feature map with the same size as the input picture, each pixel point on the feature map is a vector with the length of K4, and the sum of K4 numbers in the vector is 1, which respectively represents the probability that the pixel point is divided into each surface process mode. And taking the surface process mode with the highest probability as the surface process mode of the pixel point, and counting the number of the pixel points belonging to each surface process mode respectively to be used as a first identification result of the surface process.
After the first recognition result is obtained, the recognition result fusion can be performed based on the first recognition result to obtain a final recognition result of the jewelry characteristics, that is, a second recognition result, which specifically includes:
(e) the jewelry category: and outputting N first identification results of the jewelry sample types by the N jewelry sample pictures, counting the number of the first identification results corresponding to each jewelry type, and taking the type with the largest number of the first identification results as a second identification result of the jewelry type.
(f) Whether diamond is involved and the type of diamond: and taking all the appearing mosaic modes in the first identification results of the N jewelry sample pictures as the diamond types of the jewelry samples to obtain a second identification result of the diamond types. And if no diamond detection frame is output in the N pictures, determining that the jewelry sample has no diamond.
(g) The metal material is as follows: n jewelry sample pictures output N first identification results of the metal materials of the jewelry sample, the number of the first identification results corresponding to each metal material is counted, and the metal material with the largest number of the first identification results is taken as a second identification result of the metal materials of the jewelry sample.
(h) The surface process mode is as follows: and respectively outputting the pixel number of each surface process mode by the N jewelry sample pictures, counting the pixel number sum corresponding to each surface process mode, comparing the obtained pixel number sum with a preset threshold value, and taking a first identification result of the surface process corresponding to the pixel number sum as a second identification result of the surface process if the obtained pixel number sum exceeds the preset threshold value.
And taking the identified second identification result of the jewelry type, the second identification result of the diamond type, the second identification result of the metal material and the second identification result of the surface process as the sample characteristics of the jewelry sample.
In the embodiment, each picture is input into a pre-trained feature recognition model to obtain at least one first recognition result output by the feature recognition model, a jewelry type second recognition result is obtained according to the first recognition result of the most jewelry types, a diamond type second recognition result is obtained according to the first recognition results of all the recognized diamond types, a metal material second recognition result is obtained according to the first recognition result of the most recognized metal materials, a surface process second recognition result is obtained according to the number of pixels corresponding to each surface process first recognition result, sample features are obtained according to the jewelry type second recognition result, the diamond type second recognition result, the metal material second recognition result and the surface process second recognition result, the jewelry can be subjected to feature recognition through the jewelry picture without manually recognizing jewelry features, the inaccurate jewelry feature recognition caused by limited manual experience is avoided.
In an embodiment, the jewelry item code determining method may further include: obtaining a first jewelry identification of a first jewelry and a second jewelry identification of a second jewelry; determining a first jewelry feature vector corresponding to the first jewelry identification and a second jewelry feature vector corresponding to the second jewelry identification according to the first mapping relation; calculating a vector distance between the first jewelry feature vector and the second jewelry feature vector; and if the vector distance is smaller than the preset threshold value, judging that the first jewelry and the second jewelry are in the same money.
In the specific implementation, when it is required to judge whether two jewels are of the same type, firstly, identifiers of the two jewels, namely a first jewel identifier and a second jewel identifier, are obtained, feature vectors corresponding to the two jewel identifiers, namely a first jewel feature vector and a second jewel feature vector, are respectively determined according to a first mapping relation between the jewel identifiers and the feature vectors, a vector distance between the two jewel feature vectors is calculated, the vector distance is compared with a preset threshold, if the vector distance is smaller than the preset threshold, the two jewels can be judged to be of the same type, otherwise, if the vector distance is larger than or equal to the preset threshold, the two jewels can be judged to be of different types.
In this embodiment, by obtaining a first jewelry identifier of a first jewelry and a second jewelry identifier of a second jewelry, determining a first jewelry feature vector corresponding to the first jewelry identifier and a second jewelry feature vector corresponding to the second jewelry identifier according to the first mapping relationship, calculating a vector distance between the first jewelry feature vector and the second jewelry feature vector, if the vector distance is smaller than a preset threshold value, determining that the first jewelry and the second jewelry are of the same type, determining whether the two jewelry are of the same type through the identifiers, without manual determination, and improving the efficiency of jewelry identification of the same type.
Moreover, the method avoids inaccurate identification results caused by artificial subjective factors, and improves the accuracy of identifying the same type of jewelry.
Fig. 2 provides a schematic flow chart of another jewelry style code determination method. According to fig. 2, the jewelry code determination method may comprise the steps of:
step S210, acquiring at least one picture of a jewelry sample;
step S220, determining a sample characteristic vector according to at least one picture of the jewelry sample;
step S230, obtaining a sample identifier of the jewelry sample;
step S240, establishing a corresponding relation between the sample identifier and the sample characteristic vector to obtain the first mapping relation;
step S222, identifying sample characteristics according to at least one picture of the jewelry sample;
step S232, determining a sample money number of the jewelry sample according to the sample characteristics;
step S242, obtaining the second mapping relationship by establishing a corresponding relationship between the sample identifier and the sample money code;
step S250, acquiring a jewelry picture; the jewelry picture comprises jewelry to be identified;
step S260, obtaining jewelry feature vectors according to the jewelry pictures;
step S270, determining a jewelry identification corresponding to the jewelry to be recognized through the jewelry characteristic vector according to a preset first mapping relation; the first mapping relation is a mapping relation between the jewelry identification and the jewelry characteristic vector;
step S280, determining a jewelry money code corresponding to the jewelry to be identified through the jewelry identification according to a preset second mapping relation; the second mapping relation is the mapping relation between the jewelry identification and the jewelry money code.
It should be understood that, although the steps in the flowcharts related to the embodiments as described above are sequentially displayed as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not limited to being performed in the exact order illustrated and, unless explicitly stated herein, may be performed in other orders. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be rotated or alternated with other steps or at least a part of the steps or stages in other steps.
Based on the same inventive concept, the embodiment of the application also provides a jewelry code determining device for realizing the jewelry code determining method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the method, so specific limitations in one or more jewelry item code determination device embodiments provided below can be referred to as the limitations on the jewelry item code determination method in the foregoing, and details are not described here.
In one embodiment, as shown in fig. 3, there is provided a jewelry item code determination apparatus 300, comprising: a picture obtaining module 310, a feature vector determining module 320, an identification determining module 330, and a pattern determining module 340, wherein:
a picture acquiring module 310, configured to acquire a jewelry picture; the jewelry picture comprises jewelry to be identified;
the feature vector determining module 320 is configured to obtain a jewelry feature vector according to the jewelry picture;
the identification determining module 330 is configured to determine, according to a preset first mapping relationship, a jewelry identification corresponding to the jewelry to be identified through the jewelry feature vector; the first mapping relation is a mapping relation between the jewelry identification and the jewelry characteristic vector;
the money code determining module 340 is configured to determine, according to a preset second mapping relationship, a jewelry money code corresponding to the jewelry to be recognized through the jewelry identifier; the second mapping relation is the mapping relation between the jewelry identification and the jewelry money code.
In an embodiment, the feature vector determining module 320 is further configured to input each jewelry picture to a pre-trained feature extraction network, so as to obtain at least one picture feature vector output by the feature extraction network; and averaging the at least one picture feature vector to obtain the jewelry feature vector.
In one embodiment, the jewelry item code determining apparatus 300 further comprises:
the sample picture acquisition module is used for acquiring at least one picture of the jewelry sample;
the sample characteristic vector determining module is used for determining a sample characteristic vector according to at least one picture of the jewelry sample;
the sample identification acquisition module is used for acquiring the sample identification of the jewelry sample;
and the first mapping relation determining module is used for establishing a corresponding relation between the sample identifier and the sample feature vector to obtain the first mapping relation.
In one embodiment, the jewelry item code determining apparatus 300 further comprises:
the sample characteristic identification module is used for identifying sample characteristics according to at least one picture of the jewelry sample;
the sample money code determining module is used for determining the sample money code of the jewelry sample according to the sample characteristics;
and the second mapping relation determining module is used for establishing a corresponding relation between the sample identifier and the sample money code to obtain the second mapping relation.
In an embodiment, the sample feature recognition module is further configured to input each of the pictures into a pre-trained feature recognition model, so as to obtain at least one first recognition result output by the feature recognition model; the first identification result comprises a first identification result of jewelry type, a first identification result of diamond type, a first identification result of metal material and a first identification result of surface process; obtaining a second jewelry type identification result according to the first jewelry type identification result with the most recognized jewelry types; obtaining a second identification result of the diamond types according to the first identification results of all the identified diamond types; obtaining a second identification result of the metal material according to the first identification result of the most identified metal materials; obtaining a second recognition result of the surface process according to the pixel number corresponding to each first recognition result of the surface process; and obtaining the sample characteristics according to the second identification result of the jewelry type, the second identification result of the diamond type, the second identification result of the metal material and the second identification result of the surface process.
In one embodiment, the jewelry item code determining apparatus 300 further comprises:
the identification acquisition module is used for acquiring a first jewelry identification of the first jewelry and a second jewelry identification of the second jewelry;
the double-feature-vector determining module is used for determining a first jewelry feature vector corresponding to the first jewelry identifier and a second jewelry feature vector corresponding to the second jewelry identifier according to the first mapping relation;
a vector distance calculation module for calculating a vector distance between the first jewelry feature vector and the second jewelry feature vector;
and the judging module is used for judging that the first jewelry and the second jewelry are in the same money if the vector distance is smaller than a preset threshold value.
The various modules of the jewelry code determination apparatus described above may be implemented in whole or in part by software, hardware, and combinations thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 4. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for communicating with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a jewelry code determination method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on a shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the configuration shown in fig. 4 is a block diagram of only a portion of the configuration associated with the present application, and is not intended to limit the computing device to which the present application may be applied, and that a particular computing device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In an embodiment, a computer device is further provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above method embodiments when executing the computer program.
In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method embodiments.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above may be implemented by hardware instructions of a computer program, which may be stored in a non-volatile computer-readable storage medium, and when executed, may include the processes of the embodiments of the methods described above. Any reference to memory, databases, or other media used in the embodiments provided herein can include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), Magnetic Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), for example. The databases involved in the embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing based data processing logic devices, etc., without limitation.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, and these are all within the scope of protection of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.
Claims (8)
1. A jewelry code determining method, comprising:
obtaining at least one picture of a jewelry sample;
inputting each picture into a pre-trained feature recognition model to obtain at least one first recognition result output by the feature recognition model; the first identification result comprises a first identification result of jewelry types, a first identification result of diamond types, a first identification result of metal materials and a first identification result of surface processes;
obtaining a second jewelry type identification result according to the first jewelry type identification result with the most recognized jewelry types;
obtaining a second identification result of the diamond types according to the first identification results of all the identified diamond types;
obtaining a second identification result of the metal material according to the first identification result of the most identified metal materials;
obtaining a second recognition result of the surface process according to the pixel number corresponding to each first recognition result of the surface process;
obtaining sample characteristics of the jewelry sample according to the second identification result of the jewelry type, the second identification result of the diamond type, the second identification result of the metal material and the second identification result of the surface process; the sample features are used for determining a second mapping relation;
obtaining a jewelry picture; the jewelry picture comprises jewelry to be identified;
obtaining jewelry feature vectors according to the jewelry pictures;
determining a jewelry identifier corresponding to the jewelry to be recognized through the jewelry feature vector according to a preset first mapping relation; the first mapping relation is a mapping relation between the jewelry identification and the jewelry characteristic vector;
determining a jewelry money code corresponding to the jewelry to be identified through the jewelry identification according to the second mapping relation; the second mapping relation is the mapping relation between the jewelry identification and the jewelry money code.
2. The method of claim 1, wherein the jewelry picture is at least one; obtaining jewelry feature vectors according to the jewelry pictures, comprising:
inputting each jewelry picture into a pre-trained feature extraction network to obtain at least one picture feature vector output by the feature extraction network;
and averaging the at least one picture feature vector to obtain the jewelry feature vector.
3. The method of claim 1, further comprising:
determining a sample feature vector according to at least one picture of the jewelry sample;
obtaining a sample identification of the jewelry sample;
and establishing a corresponding relation between the sample identification and the sample characteristic vector to obtain the first mapping relation.
4. The method of claim 3, further comprising:
determining a sample money number of the jewelry sample according to the sample characteristics;
and establishing a corresponding relation between the sample identification and the sample money code to obtain the second mapping relation.
5. The method of claim 1, further comprising:
obtaining a first jewelry identification of a first jewelry and a second jewelry identification of a second jewelry;
determining a first jewelry characteristic vector corresponding to the first jewelry identification and a second jewelry characteristic vector corresponding to the second jewelry identification according to the first mapping relation;
calculating a vector distance between the first jewelry feature vector and the second jewelry feature vector;
and if the vector distance is smaller than a preset threshold value, judging that the first jewelry and the second jewelry are in the same style.
6. A jewelry code determination apparatus, comprising:
the sample picture acquisition module is used for acquiring at least one picture of the jewelry sample;
the sample feature recognition module is used for inputting each picture into a pre-trained feature recognition model to obtain at least one first recognition result output by the feature recognition model; the first identification result comprises a first identification result of jewelry types, a first identification result of diamond types, a first identification result of metal materials and a first identification result of surface processes; obtaining a second jewelry type identification result according to the first jewelry type identification result which is identified most; obtaining a second identification result of the diamond types according to the first identification results of all the identified diamond types; obtaining a second identification result of the metal material according to the first identification result of the most identified metal materials; obtaining a second recognition result of the surface process according to the pixel number corresponding to each first recognition result of the surface process; obtaining sample characteristics of the jewelry sample according to the second identification result of the jewelry type, the second identification result of the diamond type, the second identification result of the metal material and the second identification result of the surface process; the sample features are used for determining a second mapping relation;
the picture acquisition module is used for acquiring jewelry pictures; the jewelry picture comprises jewelry to be identified;
the characteristic vector determining module is used for obtaining a jewelry characteristic vector according to the jewelry picture;
the identification determining module is used for determining jewelry identification corresponding to the jewelry to be recognized through the jewelry characteristic vector according to a preset first mapping relation; the first mapping relation is a mapping relation between the jewelry identification and the jewelry characteristic vector;
the money code determining module is used for determining jewelry money codes corresponding to the jewelry to be recognized through the jewelry identification according to the second mapping relation; the second mapping relation is the mapping relation between the jewelry identification and the jewelry money code.
7. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 5.
8. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210174482.7A CN114239758B (en) | 2022-02-25 | 2022-02-25 | Jewelry money code determination method, jewelry money code determination device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210174482.7A CN114239758B (en) | 2022-02-25 | 2022-02-25 | Jewelry money code determination method, jewelry money code determination device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114239758A CN114239758A (en) | 2022-03-25 |
| CN114239758B true CN114239758B (en) | 2022-07-26 |
Family
ID=80748151
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210174482.7A Active CN114239758B (en) | 2022-02-25 | 2022-02-25 | Jewelry money code determination method, jewelry money code determination device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114239758B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115761152B (en) * | 2023-01-06 | 2023-06-23 | 深圳星坊科技有限公司 | Image processing and three-dimensional reconstruction method and device under common light source and computer equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109711927A (en) * | 2018-12-10 | 2019-05-03 | 梦工场珠宝企业管理有限公司 | Bullion method for customizing and device, display equipment and storage medium |
| CN110096627A (en) * | 2019-03-27 | 2019-08-06 | 北京星坊科技有限公司 | Querying method, device, computer equipment and the storage medium of jewelry information |
| CN113486905A (en) * | 2021-07-01 | 2021-10-08 | 深圳星坊科技有限公司 | Jewelry characteristic information acquisition method and device, computer equipment and storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2901424A1 (en) * | 2013-02-06 | 2014-08-14 | Electronics Commodities Exchange | Methods for recommending jewelry items |
| US10810647B2 (en) * | 2017-06-09 | 2020-10-20 | International Business Machines Corporation | Hybrid virtual and physical jewelry shopping experience |
| CN109886739A (en) * | 2019-01-31 | 2019-06-14 | 傲宝珠宝文化发展(深圳)有限公司 | Based on jewelry shops shopping guide's management method, system and its storage medium |
| CN113344496A (en) * | 2021-06-16 | 2021-09-03 | 国家珠宝检测中心(广东)有限责任公司 | Method and system for multi-strategy jewelry identification |
-
2022
- 2022-02-25 CN CN202210174482.7A patent/CN114239758B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109711927A (en) * | 2018-12-10 | 2019-05-03 | 梦工场珠宝企业管理有限公司 | Bullion method for customizing and device, display equipment and storage medium |
| CN110096627A (en) * | 2019-03-27 | 2019-08-06 | 北京星坊科技有限公司 | Querying method, device, computer equipment and the storage medium of jewelry information |
| CN113486905A (en) * | 2021-07-01 | 2021-10-08 | 深圳星坊科技有限公司 | Jewelry characteristic information acquisition method and device, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114239758A (en) | 2022-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10942967B1 (en) | Verifying item attributes using artificial intelligence | |
| US9349076B1 (en) | Template-based target object detection in an image | |
| CN107679475B (en) | Store monitoring and evaluating method and device and storage medium | |
| CN108292425A (en) | Automatically the image capture guided and displaying | |
| US9996763B2 (en) | Systems and methods for evaluating suitability of an article for an individual | |
| CN110363084A (en) | A kind of class state detection method, device, storage medium and electronics | |
| CN114494260A (en) | Object defect detection method and device, computer equipment and storage medium | |
| CN110287836B (en) | Image classification method and device, computer equipment and storage medium | |
| CN110580489B (en) | Data object classification system, method and equipment | |
| CN111639970A (en) | Method for determining price of article based on image recognition and related equipment | |
| CN111506758A (en) | Method and device for determining article name, computer equipment and storage medium | |
| CN112749723A (en) | Sample labeling method and device, computer equipment and storage medium | |
| CN114239758B (en) | Jewelry money code determination method, jewelry money code determination device, computer equipment and storage medium | |
| CN109544262A (en) | Item recommendation method, device, electronic equipment, system and readable storage medium storing program for executing | |
| KR20240128967A (en) | API that provides product cards | |
| CN114170435A (en) | Method and device for screening appearance images for recovery detection | |
| CN113486905A (en) | Jewelry characteristic information acquisition method and device, computer equipment and storage medium | |
| KR20240128068A (en) | Dynamically Presenting Augmented Reality Content Generators | |
| CN116206125A (en) | Appearance defect identification method, appearance defect identification device, computer equipment and storage medium | |
| CN114550175A (en) | Image processing method, image processing device, electronic equipment and computer readable storage medium | |
| CN110413869B (en) | Method and device for pushing information | |
| CN116524088A (en) | Jewelry virtual try-on method, jewelry virtual try-on device, computer equipment and storage medium | |
| CN114693414A (en) | Store item recommendation method and device, computer equipment and storage medium | |
| CN114549857A (en) | Image information identification method and device, computer equipment and storage medium | |
| CN117809009A (en) | Identity recognition method, equipment, device, computer equipment and storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |