CN112308492A - Deep learning and knowledge graph fusion-based warehouse management method and system - Google Patents
Deep learning and knowledge graph fusion-based warehouse management method and system Download PDFInfo
- Publication number
- CN112308492A CN112308492A CN202011244716.8A CN202011244716A CN112308492A CN 112308492 A CN112308492 A CN 112308492A CN 202011244716 A CN202011244716 A CN 202011244716A CN 112308492 A CN112308492 A CN 112308492A
- Authority
- CN
- China
- Prior art keywords
- warehouse
- warehouse management
- deep learning
- materials
- management robot
- 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
- 238000007726 management method Methods 0.000 title claims abstract description 177
- 238000013135 deep learning Methods 0.000 title claims abstract description 47
- 230000004927 fusion Effects 0.000 title claims abstract description 27
- 238000013136 deep learning model Methods 0.000 claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 11
- 206010063385 Intellectualisation Diseases 0.000 claims abstract description 8
- 238000001514 detection method Methods 0.000 claims abstract description 7
- 238000000605 extraction Methods 0.000 claims description 28
- 238000012360 testing method Methods 0.000 claims description 24
- 230000003993 interaction Effects 0.000 claims description 12
- 238000002372 labelling Methods 0.000 claims description 8
- 238000004519 manufacturing process Methods 0.000 claims description 7
- 230000032258 transport Effects 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
- G06Q10/0875—Itemisation or classification of parts, supplies or services, e.g. bill of materials
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
Abstract
The invention discloses a warehouse management method and system based on deep learning and knowledge map fusion, belonging to the field of warehouse management, aiming at solving the technical problems of high management intensity, low efficiency, high error rate, disordered storage and low intellectualization of the traditional warehouse at present, and adopting the technical scheme as follows: the method includes the steps that a camera is mounted on a warehouse management robot to conduct all-around real-time collection of material information, a deep learning model and a knowledge map are built according to the collected material information, then collection of in-out warehouse material information and face or industrial card identification detection are conducted, and finally high efficiency, convenience, orderliness, intellectualization and unmanned warehouse management are achieved through a cloud platform; the method comprises the following specific steps: machine deep learning; warehousing management: warehousing, registering, classifying and storing purchased materials inside a company; and (4) ex-warehouse management: and (5) carrying out warehouse-out registration and picking up of internal materials of the company. The invention also discloses a warehouse management system based on deep learning and knowledge map fusion.
Description
Technical Field
The invention relates to the field of warehouse management, in particular to a warehouse management method and a warehouse management system based on deep learning and knowledge map fusion.
Background
The advent of the big data era brings unprecedented data dividends for the rapid development of artificial intelligence. Under the condition of 'feeding' of big data, the artificial intelligence technology achieves unprecedented great progress. The progress is highlighted in the relevant fields such as knowledge engineering represented by a knowledge graph and machine learning represented by deep learning. Deep learning is the intrinsic rule and the expression level of learning sample data, the information obtained in the learning process is very helpful for explaining data such as characters, images and sounds, and the final aim is to enable a machine to have the analysis and learning capability like a human and to recognize the data such as the characters, the images and the sounds. The deep learning is a complex machine learning algorithm, the effect in the aspects of voice and image recognition is far superior to that of the prior related technology, the unprecedented data red profit is brought to the deep learning by big data, and the deep neural network can learn effective hierarchical feature representation by benefiting from large-scale labeled data, so that the excellent effect is achieved in the fields of image recognition and the like.
The knowledge graph is essentially a semantic network, expressing various entities, concepts and semantic relationships among them. Compared with the traditional knowledge representation forms (such as ontologies and traditional semantic networks), the knowledge graph has the advantages of high entity/concept coverage rate, various semantic relations, friendly structure (generally expressed in an RDF format), high quality and the like, so that the knowledge graph increasingly becomes the most main knowledge representation mode in the big data age and the artificial intelligence age.
At the present stage, the method of applying the deep learning technology to the knowledge graph is relatively direct. A large number of deep learning models can effectively complete tasks such as end-to-end entity identification, relation extraction, relation completion and the like, and further can be used for constructing or enriching a knowledge map. The combination of knowledge map and deep learning is one of the important ideas for further improving the effect of the deep learning model. The symbolic senses represented by the knowledge graph and the connection senses represented by the deep learning increasingly depart from the original independently developed tracks and walk on a new road cooperatively and further.
At present, a warehouse management system of a common medium and small company stays at a traditional employee for registering, delivering and storing, and then the employee performs classified storage treatment, so that the problems of high warehouse delivery strength, low efficiency, high error rate, disordered storage and the like exist, the production efficiency is seriously influenced, economic loss is brought to the company, and although an intelligent warehouse management system is adopted by a common large-scale company, the problems of personnel management and low intelligence still exist.
Disclosure of Invention
The technical task of the invention is to provide a warehouse management method and system based on deep learning and knowledge graph fusion, so as to solve the problems of high management intensity, low efficiency, high error rate, disordered storage and low intellectualization of the traditional warehouse at present.
The technical task of the invention is realized in the following way, the method for managing the warehouse based on deep learning and knowledge map fusion comprises the steps of carrying out all-around real-time acquisition of material information by installing a camera on a warehouse management robot, establishing a deep learning model and a knowledge map according to the acquired material information, carrying out in-and-out warehouse material information acquisition and face or work card identification detection, and finally realizing high efficiency, convenience, orderliness, intellectualization and no humanization of the warehouse management through a cloud platform; the method comprises the following specific steps:
machine deep learning: learning and knowing the information of the material types, quantity, manufacturers and specification parameters, and forming a knowledge graph through text data and a relation extraction model so as to establish the connection relation of each material;
warehousing management: warehousing, registering, classifying and storing purchased materials inside a company;
and (4) ex-warehouse management: and (5) carrying out warehouse-out registration and picking up of internal materials of the company.
Preferably, the machine deep learning is specifically as follows:
collecting material videos or pictures through a camera;
test data: acquiring a material video or picture through a camera, and preprocessing the material video or picture to form test data;
labeling data: the method comprises the steps that test data are used for producing corpus data through a data labeling platform, and then the corpus data are processed into labeled data;
a relation extraction model: inputting the labeled data into a deep learning model for training and testing to generate a relation extraction model;
entity relationship pair: inputting the text data to be extracted into a relation extraction model, and extracting an entity relation pair of the text data to be extracted;
and (5) establishing a knowledge graph, namely establishing the knowledge graph of the text to be extracted by utilizing a graph database.
Preferably, the warehousing management is specifically as follows:
starting up and initializing: when the materials need to be put in storage, for the learned material information of the warehouse management robot, company personnel need to arrive at a specified area to be put in storage and start the warehouse management robot;
identifying the sponsor: the face or work card identification of the operator is carried out, the warehousing information is convenient to trace,
operating a warehouse management robot through voice interaction or touch screen operation and automatically acquiring a complete image of a material identification area by using a camera;
the warehouse management robot outputs the material positioning frame and the class label data;
establishing a knowledge graph with the existing materials by using the established graph database;
judging whether materials wait for storage:
if yes, after the warehouse management robot confirms that the registration of the material information is finished, the warehouse management robot stores the materials in a classified mode, and therefore the materials are successfully put into the warehouse;
if not, the storage fails.
Preferably, the ex-warehouse management is as follows:
starting up and initializing: when the materials need to be taken out of the warehouse, the warehouse management robot is started through remote control or on-site acquisition;
identifying the sponsor: performing face identification or work card identification of a manager, and performing face identification or work card identification of the manager on site if the manager needs to be picked up on site;
materials to be taken are spoken through voice interaction or a camera of a warehouse management robot is controlled through a touch screen to collect material label data names;
outputting material pictures and related information by using the existing knowledge graph, and finishing related ex-warehouse registration information by using the warehouse management robot;
judging whether materials wait for being delivered from the warehouse or not:
if so, the warehouse management robot operates the warehouse to carry out the warehouse, and further the successful warehouse carrying out of the material is realized;
if not, the ex-warehouse failure is indicated.
Preferably, the library is made by the warehouse management robot, which includes the following two cases:
when the material information does not exceed the learned range of the warehouse management robot, the warehouse management robot transports related materials, the related materials are placed in the areas to be delivered in a classified mode, and a manager directly takes the materials from the areas to be delivered;
when the material information exceeds the learned range of the warehouse management robot, new material information is learned again through the deep learning model and the knowledge map, then the new material information is issued to the warehouse management robot body through the cloud platform, and then the warehouse management robot can normally finish warehouse entry and exit registration and access.
A warehouse management system based on deep learning and knowledge graph fusion comprises a hardware subsystem and a software subsystem;
wherein the hardware subsystem comprises a first hardware subsystem and a second hardware subsystem,
the warehouse management robot is used for replacing manpower to finish warehousing and ex-warehouse of materials, and is provided with a camera;
the camera is used for collecting material information in all directions in real time, establishing a deep learning model and a knowledge map, and facilitating material information collection in and out of a warehouse and human face or work card identification detection;
the cloud platform is used for remotely operating the robot and the camera;
the software sub-system comprises a software sub-system,
the machine deep learning unit is used for learning and knowing the information of the material types, the quantity, the manufacturers and the specification parameters, and forming a knowledge map through the text data and the relation extraction model so as to establish the connection relation of each material;
the warehousing management unit is used for warehousing registration and classified storage of purchased materials inside a company;
and the ex-warehouse management unit is used for ex-warehouse registration and getting of internal materials of the company.
Preferably, the machine depth learning unit includes,
the acquisition module is used for acquiring material videos or pictures through the camera;
the test data forming module is used for acquiring a material video or picture through a camera and forming test data after preprocessing;
the labeled data production module is used for producing the corpus data by using the test data through the data labeling platform and then processing the corpus data into labeled data;
the relation extraction model production module is used for inputting the labeled data into the deep learning model for training and testing to generate a relation extraction model;
the entity relationship pair extraction module is used for inputting the text data to be extracted into the relationship extraction model and extracting the entity relationship pair of the text data to be extracted;
and the knowledge map construction module is used for constructing a knowledge map of the text to be extracted by using the map database.
Preferably, the warehousing management unit includes,
the warehousing initialization module is used for enabling company personnel to reach a specified region to be warehoused for the learned material information of the warehouse management robot and starting the warehouse management robot when the materials need to be warehoused;
the warehousing identification module is used for identifying faces or work cards of workers and is convenient for tracing warehousing information,
the warehouse entry completion image acquisition module is used for operating the warehouse management robot through voice interaction or touch screen operation and automatically acquiring a complete image of the material identification area by using a camera;
the warehouse entry output module is used for outputting the material positioning frame and the class label data through the warehouse management robot;
the warehousing knowledge map building module is used for building a knowledge map by using the established map database and the existing materials;
the warehousing judgment module is used for judging whether materials wait for warehousing:
if yes, after the warehouse management robot confirms that the registration of the material information is finished, the warehouse management robot stores the materials in a classified mode, and therefore the materials are successfully put into the warehouse;
if not, the storage fails.
Preferably, the ex-warehouse management unit includes,
the ex-warehouse initialization module is used for starting the warehouse management robot through remote control or on-site acquisition when the materials need to be ex-warehouse;
the ex-warehouse identification module is used for carrying out face identification or work card identification of a manager, and if the on-site acquisition is carried out, face identification or work card identification of the manager is required to be carried out on site;
the warehouse-out material information acquisition module is used for speaking out the materials to be taken through voice interaction or controlling a warehouse management robot camera to acquire material label data names through a touch screen;
the ex-warehouse registration module is used for outputting material pictures and related information by using the existing knowledge graph, and the warehouse management robot finishes the related ex-warehouse registration information;
the ex-warehouse judging module is used for judging whether materials wait for ex-warehouse:
if so, the warehouse management robot operates the warehouse to carry out the warehouse, and further the successful warehouse carrying out of the material is realized; the library operation by the warehouse management robot includes the following two cases:
when the material information does not exceed the learned range of the warehouse management robot, the warehouse management robot transports related materials, the related materials are placed in the areas to be delivered in a classified mode, and a manager directly takes the materials from the areas to be delivered;
and secondly, when the material information exceeds the learned range of the warehouse management robot, learning new material information again through the deep learning model and the knowledge graph, issuing the new material information to the warehouse management robot body through the cloud platform, and normally finishing warehouse entry and exit registration and access by the warehouse management robot.
If not, the ex-warehouse failure is indicated.
A computer readable storage medium having stored thereon computer executable instructions, which when executed by a processor, implement a deep learning and knowledge-graph fusion based warehouse management method as described above.
The warehouse management method and system based on deep learning and knowledge graph fusion have the following advantages:
the invention solves the problems of high management intensity, low efficiency, high error rate, disordered storage, low intellectualization and the like of the traditional warehouse at present, combines deep learning and a knowledge map, and enables the material management of a company to be more efficient, convenient, organized, intelligent and unmanned through a cloud platform;
the robot can realize management and access of the warehouse as a person, and can be replaced by one robot from 2 to 3 original warehouse managers, so that the personnel cost is reduced;
the invention can complete material identification and classification by identifying picture or video information, thereby improving the warehousing efficiency, and simultaneously completing ordered storage and pickup, thereby avoiding the problems of disordered warehouse materials, material loss and material leakage;
the invention can continuously learn new material information, and can establish the relationship of each material, for example, a set of machines is assembled, when the material is delivered out of the warehouse, the material list of the whole set of machines does not need to be clearly listed, and only the names of the machines needing to be assembled are listed, the warehouse management system can transport all the materials with the related required quantity to the region where the material is to be delivered out of the warehouse, thereby improving the delivery efficiency and realizing the intelligent operation of warehouse management;
the invention can be applied to the technical field of warehouse management, reduces personnel and management cost of companies, improves the warehouse entry and exit efficiency, reduces the labor intensity, realizes intellectualization and unmanned warehouse management, and has better application prospect.
Drawings
The invention is further described below with reference to the accompanying drawings.
FIG. 1 is a flow chart diagram of a warehouse management method based on deep learning and knowledge graph fusion;
FIG. 2 is a schematic diagram of the operation of a warehouse management method based on deep learning and knowledge graph fusion;
FIG. 3 is a block flow diagram of machine deep learning;
FIG. 4 is a block diagram of a process for warehousing management;
fig. 5 is a block diagram of the flow of ex-warehouse management.
Detailed Description
The deep learning and knowledge graph fusion-based warehouse management method and system of the present invention are described in detail below with reference to the drawings and specific embodiments of the specification.
Example 1:
the invention relates to a warehouse management method based on deep learning and knowledge map fusion, which is characterized in that a camera is arranged on a warehouse management robot to acquire material information in all directions in real time, a deep learning model and a knowledge map are established according to the acquired material information, then the information acquisition of the material entering and exiting the warehouse and the identification and detection of human faces or work cards are carried out, and finally the high efficiency, convenience, orderliness, intellectualization and no humanization of the warehouse management are realized through a cloud platform; the warehouse management robot is provided with 2-4 cameras; as shown in fig. 1, the following are specific:
s1, machine deep learning: learning and knowing the information of the material types, quantity, manufacturers and specification parameters, and forming a knowledge graph through text data and a relation extraction model so as to establish the connection relation of each material;
s2, warehousing management: warehousing, registering, classifying and storing purchased materials inside a company;
s3, warehouse-out management: and (5) carrying out warehouse-out registration and picking up of internal materials of the company.
As shown in fig. 3, the machine deep learning in step S1 of the present embodiment is specifically as follows:
s101, collecting material videos or pictures through a camera;
s102, test data: acquiring a material video or picture through a camera, and preprocessing the material video or picture to form test data;
s103, marking data: the method comprises the steps that test data are used for producing corpus data through a data labeling platform, and then the corpus data are processed into labeled data;
s104, a relation extraction model: inputting the labeled data into a deep learning model for training and testing to generate a relation extraction model;
s105, entity relationship pair: inputting the text data to be extracted into a relation extraction model, and extracting an entity relation pair of the text data to be extracted;
and S106, constructing a knowledge graph, namely constructing the knowledge graph of the text to be extracted by utilizing the graph database.
As shown in fig. 4, the warehousing management in step S2 of this embodiment is as follows:
s201, starting initialization: when the materials need to be put in storage, for the learned material information of the warehouse management robot, company personnel need to arrive at a specified area to be put in storage and start the warehouse management robot;
s202, identifying the sponsor: the face or work card identification of the operator is carried out, the warehousing information is convenient to trace,
s203, operating the warehouse management robot through voice interaction or touch screen operation and automatically acquiring a complete image of the material identification area by using a camera;
s204, the warehouse management robot outputs the material positioning frame and the class label data;
s205, constructing a knowledge graph with the existing materials by using the established graph database;
s206, judging whether materials wait for storage:
if yes, after confirming that the material information is registered, the warehouse management robot classifies and stores the materials so as to successfully put the materials into a warehouse;
and if not, the storage fails.
As shown in fig. 5, the ex-warehouse management in step S3 of the present embodiment is as follows:
s301, starting initialization: when the materials need to be taken out of the warehouse, the warehouse management robot is started through remote control or on-site acquisition;
s302, identifying the sponsor: performing face identification or work card identification of a manager, and performing face identification or work card identification of the manager on site if the manager needs to be picked up on site;
s303, speaking the materials to be taken through voice interaction or controlling a camera of the warehouse management robot to collect the material label data name through a touch screen;
s304, outputting the material picture and related information by using the existing knowledge graph, and finishing related ex-warehouse registration information by the warehouse management robot;
s305, judging whether materials wait for being delivered from the warehouse:
if yes, the warehouse management robot operates the warehouse to go out, and then the successful material out of the warehouse is realized;
and if not, indicating that the warehouse-out fails.
Wherein, the warehouse management robot operating to make the warehouse comprises the following two conditions:
when the material information does not exceed the learned range of the warehouse management robot, the warehouse management robot transports related materials, the related materials are placed in the areas to be delivered in a classified mode, and a manager directly takes the materials from the areas to be delivered;
when the material information exceeds the learned range of the warehouse management robot, new material information is learned again through the deep learning model and the knowledge map, then the new material information is issued to the warehouse management robot body through the cloud platform, and then the warehouse management robot can normally finish warehouse entry and exit registration and access.
Example 2:
the invention relates to a deep learning and knowledge graph fusion-based warehouse management system, which comprises a hardware subsystem and a software subsystem;
wherein the hardware subsystem comprises a first hardware subsystem and a second hardware subsystem,
the warehouse management robot is used for replacing manpower to finish warehousing and ex-warehouse of materials, and is provided with a camera;
the camera is used for collecting material information in all directions in real time, establishing a deep learning model and a knowledge map, and facilitating material information collection in and out of a warehouse and human face or work card identification detection;
the cloud platform is used for remotely operating the robot and the camera;
the software sub-system comprises a software sub-system,
the machine deep learning unit is used for learning and knowing the information of the material types, the quantity, the manufacturers and the specification parameters, and forming a knowledge map through the text data and the relation extraction model so as to establish the connection relation of each material;
the warehousing management unit is used for warehousing registration and classified storage of purchased materials inside a company;
and the ex-warehouse management unit is used for ex-warehouse registration and getting of internal materials of the company.
The machine depth learning unit in the present embodiment includes,
the acquisition module is used for acquiring material videos or pictures through the camera;
the test data forming module is used for acquiring a material video or picture through a camera and forming test data after preprocessing;
the labeled data production module is used for producing the corpus data by using the test data through the data labeling platform and then processing the corpus data into labeled data;
the relation extraction model production module is used for inputting the labeled data into the deep learning model for training and testing to generate a relation extraction model;
the entity relationship pair extraction module is used for inputting the text data to be extracted into the relationship extraction model and extracting the entity relationship pair of the text data to be extracted;
and the knowledge map construction module is used for constructing a knowledge map of the text to be extracted by using the map database.
The warehousing management unit in the present embodiment includes,
the warehousing initialization module is used for enabling company personnel to reach a specified region to be warehoused for the learned material information of the warehouse management robot and starting the warehouse management robot when the materials need to be warehoused;
the warehousing identification module is used for identifying faces or work cards of workers and is convenient for tracing warehousing information,
the warehouse entry completion image acquisition module is used for operating the warehouse management robot through voice interaction or touch screen operation and automatically acquiring a complete image of the material identification area by using a camera;
the warehouse entry output module is used for outputting the material positioning frame and the class label data through the warehouse management robot;
the warehousing knowledge map building module is used for building a knowledge map by using the established map database and the existing materials;
the warehousing judgment module is used for judging whether materials wait for warehousing:
if yes, after the warehouse management robot confirms that the registration of the material information is finished, the warehouse management robot stores the materials in a classified mode, and therefore the materials are successfully put into the warehouse;
if not, the storage fails.
The ex-warehouse management unit in the present embodiment includes,
the ex-warehouse initialization module is used for starting the warehouse management robot through remote control or on-site acquisition when the materials need to be ex-warehouse;
the ex-warehouse identification module is used for carrying out face identification or work card identification of a manager, and if the on-site acquisition is carried out, face identification or work card identification of the manager is required to be carried out on site;
the warehouse-out material information acquisition module is used for speaking out the materials to be taken through voice interaction or controlling a warehouse management robot camera to acquire material label data names through a touch screen;
the ex-warehouse registration module is used for outputting material pictures and related information by using the existing knowledge graph, and the warehouse management robot finishes the related ex-warehouse registration information;
the ex-warehouse judging module is used for judging whether materials wait for ex-warehouse:
if so, the warehouse management robot operates the warehouse to carry out the warehouse, and further the successful warehouse carrying out of the material is realized; the library operation by the warehouse management robot includes the following two cases:
when the material information does not exceed the learned range of the warehouse management robot, the warehouse management robot transports related materials, the related materials are placed in the areas to be delivered in a classified mode, and a manager directly takes the materials from the areas to be delivered;
and secondly, when the material information exceeds the learned range of the warehouse management robot, learning new material information again through the deep learning model and the knowledge graph, issuing the new material information to the warehouse management robot body through the cloud platform, and normally finishing warehouse entry and exit registration and access by the warehouse management robot.
If not, the ex-warehouse failure is indicated.
Example 3:
the embodiment of the invention also provides a computer-readable storage medium, wherein a plurality of instructions are stored, and the instructions are loaded by the processor, so that the processor executes the warehouse management method based on deep learning and knowledge graph fusion in any embodiment of the invention. Specifically, a system or an apparatus equipped with a storage medium on which software program codes that realize the functions of any of the above-described embodiments are stored may be provided, and a computer (or a CPU or MPU) of the system or the apparatus is caused to read out and execute the program codes stored in the storage medium.
In this case, the program code itself read from the storage medium can realize the functions of any of the above-described embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.
Examples of the storage medium for supplying the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (e.g., CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-R, a warehouse management method and system based on deep learning and knowledge map fusion M, DVD-RW, DVD + RW), a magnetic tape, a nonvolatile memory card, and a ROM. Alternatively, the program code may be downloaded from a server computer via a communications network.
Further, it should be clear that the functions of any one of the above-described embodiments may be implemented not only by executing the program code read out by the computer, but also by causing an operating system or the like operating on the computer to perform a part or all of the actual operations based on instructions of the program code.
Further, it is to be understood that the program code read out from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion unit connected to the computer, and then causes a CPU or the like mounted on the expansion board or the expansion unit to perform part or all of the actual operations based on instructions of the program code, thereby realizing the functions of any of the above-described embodiments.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A warehouse management method based on deep learning and knowledge map fusion is characterized in that a camera is mounted on a warehouse management robot to acquire material information in all directions in real time, a deep learning model and a knowledge map are established according to the acquired material information, then in-out warehouse material information acquisition and face or industrial card identification detection are performed, and finally high efficiency, convenience, orderliness, intellectualization and no humanization of warehouse management are realized through a cloud platform; the method comprises the following specific steps:
machine deep learning: learning and knowing the information of the material types, quantity, manufacturers and specification parameters, and forming a knowledge graph through text data and a relation extraction model so as to establish the connection relation of each material;
warehousing management: warehousing, registering, classifying and storing purchased materials inside a company;
and (4) ex-warehouse management: and (5) carrying out warehouse-out registration and picking up of internal materials of the company.
2. The deep learning and knowledge graph fusion based warehouse management method according to claim 1, characterized in that machine deep learning specifically comprises the following steps:
collecting material videos or pictures through a camera;
test data: acquiring a material video or picture through a camera, and preprocessing the material video or picture to form test data;
labeling data: the method comprises the steps that test data are used for producing corpus data through a data labeling platform, and then the corpus data are processed into labeled data;
a relation extraction model: inputting the labeled data into a deep learning model for training and testing to generate a relation extraction model;
entity relationship pair: inputting the text data to be extracted into a relation extraction model, and extracting an entity relation pair of the text data to be extracted;
and (5) establishing a knowledge graph, namely establishing the knowledge graph of the text to be extracted by utilizing a graph database.
3. The deep learning and knowledge graph fusion-based warehouse management method according to claim 1, characterized in that the warehouse management specifically comprises:
starting up and initializing: when the materials need to be put in storage, for the learned material information of the warehouse management robot, company personnel need to arrive at a specified area to be put in storage and start the warehouse management robot;
identifying the sponsor: the face or work card identification of the operator is carried out, the warehousing information is convenient to trace,
operating a warehouse management robot through voice interaction or touch screen operation and automatically acquiring a complete image of a material identification area by using a camera;
the warehouse management robot outputs the material positioning frame and the class label data;
establishing a knowledge graph with the existing materials by using the established graph database;
judging whether materials wait for storage:
if yes, after the warehouse management robot confirms that the registration of the material information is finished, the warehouse management robot stores the materials in a classified mode, and therefore the materials are successfully put into the warehouse;
if not, the storage fails.
4. The deep learning and knowledge graph fusion-based warehouse management method according to any one of claims 1 to 3, characterized in that ex-warehouse management specifically comprises the following steps:
starting up and initializing: when the materials need to be taken out of the warehouse, the warehouse management robot is started through remote control or on-site acquisition;
identifying the sponsor: performing face identification or work card identification of a manager, and performing face identification or work card identification of the manager on site if the manager needs to be picked up on site;
materials to be taken are spoken through voice interaction or a camera of a warehouse management robot is controlled through a touch screen to collect material label data names;
outputting material pictures and related information by using the existing knowledge graph, and finishing related ex-warehouse registration information by using the warehouse management robot;
judging whether materials wait for being delivered from the warehouse or not:
if so, the warehouse management robot operates the warehouse to carry out the warehouse, and further the successful warehouse carrying out of the material is realized;
if not, the ex-warehouse failure is indicated.
5. The deep learning and knowledge-graph fusion based warehouse management method according to claim 4, wherein the making of the library by the warehouse management robot comprises the following two cases:
when the material information does not exceed the learned range of the warehouse management robot, the warehouse management robot transports related materials, the related materials are placed in the areas to be delivered in a classified mode, and a manager directly takes the materials from the areas to be delivered;
when the material information exceeds the learned range of the warehouse management robot, new material information is learned again through the deep learning model and the knowledge map, then the new material information is issued to the warehouse management robot body through the cloud platform, and then the warehouse management robot can normally finish warehouse entry and exit registration and access.
6. A warehouse management system based on deep learning and knowledge graph fusion is characterized by comprising a hardware subsystem and a software subsystem;
wherein the hardware subsystem comprises a first hardware subsystem and a second hardware subsystem,
the warehouse management robot is used for replacing manpower to finish warehousing and ex-warehouse of materials, and is provided with a camera;
the camera is used for collecting material information in all directions in real time, establishing a deep learning model and a knowledge map, and facilitating material information collection in and out of a warehouse and human face or work card identification detection;
the cloud platform is used for remotely operating the robot and the camera;
the software sub-system comprises a software sub-system,
the machine deep learning unit is used for learning and knowing the information of the material types, the quantity, the manufacturers and the specification parameters, and forming a knowledge map through the text data and the relation extraction model so as to establish the connection relation of each material;
the warehousing management unit is used for warehousing registration and classified storage of purchased materials inside a company;
and the ex-warehouse management unit is used for ex-warehouse registration and getting of internal materials of the company.
7. The deep learning and knowledge-graph fusion based warehouse management system of claim 6, wherein the machine deep learning unit comprises,
the acquisition module is used for acquiring material videos or pictures through the camera;
the test data forming module is used for acquiring a material video or picture through a camera and forming test data after preprocessing;
the labeled data production module is used for producing the corpus data by using the test data through the data labeling platform and then processing the corpus data into labeled data;
the relation extraction model production module is used for inputting the labeled data into the deep learning model for training and testing to generate a relation extraction model;
the entity relationship pair extraction module is used for inputting the text data to be extracted into the relationship extraction model and extracting the entity relationship pair of the text data to be extracted;
and the knowledge map construction module is used for constructing a knowledge map of the text to be extracted by using the map database.
8. The deep learning and knowledge-graph fusion based warehouse management system of claim 6, wherein the warehousing management unit comprises,
the warehousing initialization module is used for enabling company personnel to reach a specified region to be warehoused for the learned material information of the warehouse management robot and starting the warehouse management robot when the materials need to be warehoused;
the warehousing identification module is used for identifying faces or work cards of workers and is convenient for tracing warehousing information,
the warehouse entry completion image acquisition module is used for operating the warehouse management robot through voice interaction or touch screen operation and automatically acquiring a complete image of the material identification area by using a camera;
the warehouse entry output module is used for outputting the material positioning frame and the class label data through the warehouse management robot;
the warehousing knowledge map building module is used for building a knowledge map by using the established map database and the existing materials;
the warehousing judgment module is used for judging whether materials wait for warehousing:
if yes, after the warehouse management robot confirms that the registration of the material information is finished, the warehouse management robot stores the materials in a classified mode, and therefore the materials are successfully put into the warehouse;
if not, the storage fails.
9. The deep learning and knowledge-graph fusion based warehouse management system according to any one of claims 6 to 8, wherein the ex-warehouse management unit comprises,
the ex-warehouse initialization module is used for starting the warehouse management robot through remote control or on-site acquisition when the materials need to be ex-warehouse;
the ex-warehouse identification module is used for carrying out face identification or work card identification of a manager, and if the on-site acquisition is carried out, face identification or work card identification of the manager is required to be carried out on site;
the warehouse-out material information acquisition module is used for speaking out the materials to be taken through voice interaction or controlling a warehouse management robot camera to acquire material label data names through a touch screen;
the ex-warehouse registration module is used for outputting material pictures and related information by using the existing knowledge graph, and the warehouse management robot finishes the related ex-warehouse registration information;
the ex-warehouse judging module is used for judging whether materials wait for ex-warehouse:
if so, the warehouse management robot operates the warehouse to carry out the warehouse, and further the successful warehouse carrying out of the material is realized; the library operation by the warehouse management robot includes the following two cases:
when the material information does not exceed the learned range of the warehouse management robot, the warehouse management robot transports related materials, the related materials are placed in the areas to be delivered in a classified mode, and a manager directly takes the materials from the areas to be delivered;
and secondly, when the material information exceeds the learned range of the warehouse management robot, learning new material information again through the deep learning model and the knowledge graph, issuing the new material information to the warehouse management robot body through the cloud platform, and normally finishing warehouse entry and exit registration and access by the warehouse management robot.
If not, the ex-warehouse failure is indicated.
10. A computer-readable storage medium having stored thereon computer-executable instructions that, when executed by a processor, perform a deep learning and knowledge-graph fusion based warehouse management method according to claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011244716.8A CN112308492A (en) | 2020-11-10 | 2020-11-10 | Deep learning and knowledge graph fusion-based warehouse management method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011244716.8A CN112308492A (en) | 2020-11-10 | 2020-11-10 | Deep learning and knowledge graph fusion-based warehouse management method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112308492A true CN112308492A (en) | 2021-02-02 |
Family
ID=74325540
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011244716.8A Pending CN112308492A (en) | 2020-11-10 | 2020-11-10 | Deep learning and knowledge graph fusion-based warehouse management method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112308492A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111861311A (en) * | 2020-06-24 | 2020-10-30 | 山东临工工程机械有限公司 | Machine type automatic identification real-time inventory statistics management method capable of realizing warehousing inspection |
| CN114782761A (en) * | 2022-06-23 | 2022-07-22 | 山东能源数智云科技有限公司 | Intelligent storage material identification method and system based on deep learning |
| CN117151338A (en) * | 2023-09-08 | 2023-12-01 | 安徽大学 | Multi-unmanned aerial vehicle task planning method based on large language model |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106516530A (en) * | 2016-12-21 | 2017-03-22 | 锥能机器人（上海）有限公司 | Automatic warehousing sorting station and sorting method thereof |
| CN107077659A (en) * | 2016-09-26 | 2017-08-18 | 达闼科技（北京）有限公司 | A kind of intelligent inventory management system, server, method, terminal and program product |
| CN107705066A (en) * | 2017-09-15 | 2018-02-16 | 广州唯品会研究院有限公司 | Information input method and electronic equipment during a kind of commodity storage |
| CN110377755A (en) * | 2019-07-03 | 2019-10-25 | 江苏省人民医院（南京医科大学第一附属医院） | Reasonable medication knowledge map construction method based on medicine specification |
| CN110598000A (en) * | 2019-08-01 | 2019-12-20 | 达而观信息科技（上海）有限公司 | Relationship extraction and knowledge graph construction method based on deep learning model |
| CN110825882A (en) * | 2019-10-09 | 2020-02-21 | 西安交通大学 | Knowledge graph-based information system management method |
| CN110910065A (en) * | 2019-11-21 | 2020-03-24 | 秒针信息技术有限公司 | Warehouse space distribution method and system based on big data and knowledge graph |
| CN111126888A (en) * | 2018-10-31 | 2020-05-08 | 南京智能仿真技术研究院有限公司 | Warehouse management system with high automation degree |
| CN111386233A (en) * | 2018-03-27 | 2020-07-07 | 株式会社日立产业机器 | Warehouse system |
| CN111476520A (en) * | 2020-04-03 | 2020-07-31 | 上海明略人工智能(集团)有限公司 | Method and device for determining placement position, storage medium and electronic device |
-
2020
- 2020-11-10 CN CN202011244716.8A patent/CN112308492A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107077659A (en) * | 2016-09-26 | 2017-08-18 | 达闼科技（北京）有限公司 | A kind of intelligent inventory management system, server, method, terminal and program product |
| CN106516530A (en) * | 2016-12-21 | 2017-03-22 | 锥能机器人（上海）有限公司 | Automatic warehousing sorting station and sorting method thereof |
| CN107705066A (en) * | 2017-09-15 | 2018-02-16 | 广州唯品会研究院有限公司 | Information input method and electronic equipment during a kind of commodity storage |
| CN111386233A (en) * | 2018-03-27 | 2020-07-07 | 株式会社日立产业机器 | Warehouse system |
| CN111126888A (en) * | 2018-10-31 | 2020-05-08 | 南京智能仿真技术研究院有限公司 | Warehouse management system with high automation degree |
| CN110377755A (en) * | 2019-07-03 | 2019-10-25 | 江苏省人民医院（南京医科大学第一附属医院） | Reasonable medication knowledge map construction method based on medicine specification |
| CN110598000A (en) * | 2019-08-01 | 2019-12-20 | 达而观信息科技（上海）有限公司 | Relationship extraction and knowledge graph construction method based on deep learning model |
| CN110825882A (en) * | 2019-10-09 | 2020-02-21 | 西安交通大学 | Knowledge graph-based information system management method |
| CN110910065A (en) * | 2019-11-21 | 2020-03-24 | 秒针信息技术有限公司 | Warehouse space distribution method and system based on big data and knowledge graph |
| CN111476520A (en) * | 2020-04-03 | 2020-07-31 | 上海明略人工智能(集团)有限公司 | Method and device for determining placement position, storage medium and electronic device |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111861311A (en) * | 2020-06-24 | 2020-10-30 | 山东临工工程机械有限公司 | Machine type automatic identification real-time inventory statistics management method capable of realizing warehousing inspection |
| CN114782761A (en) * | 2022-06-23 | 2022-07-22 | 山东能源数智云科技有限公司 | Intelligent storage material identification method and system based on deep learning |
| CN117151338A (en) * | 2023-09-08 | 2023-12-01 | 安徽大学 | Multi-unmanned aerial vehicle task planning method based on large language model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112308492A (en) | Deep learning and knowledge graph fusion-based warehouse management method and system | |
| CN115409069A (en) | Village and town building identification method, classification method, device, electronic equipment and medium | |
| CN110931112A (en) | Brain medical image analysis method based on multi-dimensional information fusion and deep learning | |
| CN110428069A (en) | Electric power instrument intelligent management, system, equipment and medium | |
| CN112257740B (en) | Knowledge graph-based image hidden danger identification method and system | |
| CN110781381A (en) | Data verification method, device and equipment based on neural network and storage medium | |
| KR102543064B1 (en) | System for providing manufacturing environment monitoring service based on robotic process automation | |
| CN109064467A (en) | Analysis method, device and the electronic equipment of community security defence | |
| CN113674216A (en) | Subway tunnel disease detection method based on deep learning | |
| CN117093260B (en) | Fusion model website structure analysis method based on decision tree classification algorithm | |
| CN117437647A (en) | Oracle character detection method based on deep learning and computer vision | |
| CN109242431B (en) | Enterprise management method and system based on data system | |
| CN115546824B (en) | Taboo picture identification method, apparatus and storage medium | |
| Biffl et al. | Building Empirical Software Engineering Bodies of Knowledge with Systematic Knowledge Engineering. | |
| US11580666B2 (en) | Localization and mapping method and moving apparatus | |
| Mostafa et al. | Application of Artificial Intelligence Tools with BIM Technology in Construction Management: Literature | |
| CN115984158A (en) | Defect analysis method and device, electronic equipment and computer readable storage medium | |
| CN112558512A (en) | Intelligent control and application system based on big data and Internet of things technology | |
| CN114034260A (en) | Deep foundation pit support structure deformation diagnosis system based on streaming media and BIM | |
| CN113643163A (en) | Internet education student comprehensive portrait label management system based on deep learning | |
| CN112968941B (en) | Data acquisition and man-machine collaborative annotation method based on edge calculation | |
| CN115565201B (en) | Taboo picture identification method, apparatus and storage medium | |
| CN114356502B (en) | Unstructured data marking, training and publishing system and method based on edge computing technology | |
| Ivaschenko et al. | Prelaunch matching architecture for distributed intelligent image recognition | |
| Lika et al. | Lightweight Deep Learning for Object Detection on Mobile Device |
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: 20210202 |