CN117315557A - Image recognition and classification method based on neural network - Google Patents

Image recognition and classification method based on neural network Download PDF

Info

Publication number
CN117315557A
CN117315557A CN202311340461.9A CN202311340461A CN117315557A CN 117315557 A CN117315557 A CN 117315557A CN 202311340461 A CN202311340461 A CN 202311340461A CN 117315557 A CN117315557 A CN 117315557A
Authority
CN
China
Prior art keywords
product
products
image
shooting
product appearance
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
Application number
CN202311340461.9A
Other languages
Chinese (zh)
Inventor
吴松华
张莹
周莉婷
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Saiyinsi Culture Technology Group Co ltd
Original Assignee
Shanghai Saiyinsi Culture Technology Group Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Saiyinsi Culture Technology Group Co ltd filed Critical Shanghai Saiyinsi Culture Technology Group Co ltd
Priority to CN202311340461.9A priority Critical patent/CN117315557A/en
Publication of CN117315557A publication Critical patent/CN117315557A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • 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
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Abstract

The invention discloses an image recognition and classification method based on a neural network, which belongs to the technical field of image processing and comprises the following steps: s1: training a model; s2: warehousing management; s3: and (5) warehouse-out management. The invention can identify and classify the products in the product appearance image at the single outlet/inlet by utilizing the trained neural network model, further acquire the product quantity and the category information in the product appearance image at the single outlet/inlet, conveniently update the real-time inventory quantity of various products based on the product quantity and the category information, does not need to manufacture two-dimensional codes/bar codes in advance, has relatively simple operation, and reduces the cost of product in-out warehouse management to a certain extent.

Description

Image recognition and classification method based on neural network
Technical Field
The invention relates to the technical field of image recognition, in particular to an image recognition and classification method based on a neural network.
Background
The warehouse is used for storing and safeguarding commodities and articles. The warehouse is a general name of a building and a place for storing, safeguarding and storing articles, and can be a house building, a cave, a large container or a specific place, and the like, and has the functions of storing and protecting articles. The term "store" means store and reserve, and means store for use, store, and deliver. The warehouse is a comprehensive place for reflecting the material activity condition of factories and is a transfer station for connecting production, supply and sales, and plays an important auxiliary role in promoting production and improving efficiency. The storage is to temporarily store the products and the articles due to the prepositioning of orders or the prepositioning of market prediction in the production and circulation process of the products. Meanwhile, the storage entity is surrounded to move, and clear and accurate report forms, bill accounts and accurate information calculated by accounting departments are also carried out at the same time, so that the storage is the integration of logistics, information flow and document flow. The traditional warehouse is to utilize warehouse to carry out the warehouse entry, storage and delivery activities of articles to various materials and related facility equipment. Modern warehousing refers to adding links of in-warehouse processing, sorting, in-warehouse packaging and the like on the basis of traditional warehousing. Storage is one of the important links of production and manufacture and commodity circulation, and is also an important link of logistics activities.
In the prior warehouse management process, products are mainly classified and identified by adopting a mode of scanning two-dimensional codes/bar codes, the two-dimensional codes/bar codes are required to be manufactured in advance, the operation is relatively complex, and the warehouse management cost of the products is improved to a certain extent. Therefore, an image recognition and classification method based on a neural network is provided and is used for the warehouse-in and warehouse-out management work of products.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: the method solves the problems that in the prior art, when products are put in and put out, the two-dimensional codes/bar codes are scanned to classify and identify the products, the two-dimensional codes/bar codes are required to be manufactured in advance, the operation is relatively complicated, and the cost of the product in and put out management is improved to a certain extent.
The invention solves the technical problems through the following technical proposal, and the invention comprises the following steps:
s1: model training
Training the neural network through the sample image added with the class label to obtain a product identification classification model;
s2: warehouse entry management
Receiving a warehousing shooting instruction sent by an upper computer, shooting a plurality of products at an entrance by using a first industrial camera for one time to obtain a product appearance image, then obtaining the quantity and type information of the products in the product appearance image through a product identification classification model, and updating the real-time inventory quantity of each type of products according to the quantity and type information of the products in the product appearance image at the entrance;
s3: exit management
And receiving an ex-warehouse shooting instruction sent by the upper computer, shooting a plurality of products at the outlet by using a second industrial camera for a single time to obtain a product appearance image, then obtaining the quantity and type information of the products in the product appearance image through a product identification classification model, and updating the real-time inventory quantity of each type of product according to the quantity and type information of the products in the product appearance image at the outlet.
Further, in the step S1, the training process of the product identification classification model is as follows:
s11: adding category labels to sample images of various products in a manual labeling mode to obtain a plurality of sample images with category labels, and forming a data set;
s12: dividing a sample image in a data set into a training set and a testing set according to the proportion of 3:7;
s13: inputting the sample images in the training set into a selected neural network, and training the network to obtain a trained network model;
s14: and (3) inputting the sample images in the test set into the trained network model in the step (S13), and when the performance index of the trained network model reaches a set value, storing parameters of the network model to obtain the product identification classification model.
Still further, in the step S13, the neural network includes an SSD network and a yolv 3 network.
Further, in the step S2, the specific process of acquiring the number and the category information of the products in the product appearance image through the product identification classification model is as follows:
s21: acquiring a product appearance image at the current entrance, and classifying products in the image through a product identification classification model;
s22: and counting the number information and the category information of the target detection frames of the product, namely acquiring the number and the category information of the product.
Further, in the step S3, the specific process of acquiring the number and the category information of the products in the product appearance image through the product identification classification model is as follows:
s31: acquiring a product appearance image at the current outlet, and classifying products in the image through a product identification classification model;
s32: and counting the number information and the category information of the target detection frames of the product, namely acquiring the number and the category information of the product.
Furthermore, the image recognition and classification method based on the neural network realizes the warehouse-in and warehouse-out management work of products by utilizing an image recognition and classification system based on the neural network, wherein the image recognition and classification system based on the neural network comprises a model training module, a warehouse-in shooting recognition module, a warehouse-out shooting recognition module and a database management module;
the model training module is used for training the neural network through the sample image added with the class label to obtain a product identification classification model;
the system comprises a database management module, a product appearance image acquisition module, a product identification and classification module, a product appearance image acquisition module and a database management module, wherein the database management module is used for acquiring a product appearance image of a plurality of products at an entrance by utilizing a first industrial camera, acquiring the quantity and class information of the products in the product appearance image by utilizing a product identification and classification model, and transmitting the quantity and class information of the products in the product appearance image at the entrance;
the ex-warehouse shooting identification module is used for receiving an ex-warehouse shooting instruction sent by the upper computer, shooting a plurality of products at the outlet by using the second industrial camera for a single time to obtain a product appearance image, then obtaining the quantity and type information of the products in the product appearance image through the product identification classification model, and sending the quantity and type information of the products in the product appearance image at the outlet to the database management module;
the database management module is used for receiving the quantity and the category information of the products at the single entry/exit, and updating the real-time inventory quantity of various products according to the quantity and the category information of the products at the single entry/exit.
Further, the warehousing shooting identification module comprises a warehousing shooting instruction receiving unit, a first product shooting unit and a first image identification and classification unit; the storage shooting instruction receiving unit is used for receiving a storage shooting instruction sent by the upper computer and sending the instruction to the first product shooting unit; the first product shooting unit is used for shooting a plurality of products at the entrance by using a first industrial camera at the entrance according to the warehousing shooting instruction, obtaining a product appearance image, and sending the product appearance image to the first image identification and classification unit; the first image recognition and classification unit is used for acquiring the quantity and class information of the products in the product appearance image by utilizing the product recognition and classification model, and sending the quantity and class information of the products in the product appearance image at the current entrance to the database management module.
Further, the ex-warehouse shooting identification module comprises an ex-warehouse shooting instruction receiving unit, a second product shooting unit and a second image identification and classification unit; the ex-warehouse shooting instruction receiving unit is used for receiving an ex-warehouse shooting instruction sent by the upper computer and sending the instruction to the second product shooting unit; the second product shooting unit is used for shooting a plurality of products at the outlet by using a second industrial camera at the outlet according to the ex-warehouse shooting instruction, obtaining product appearance images and sending the product appearance images to the second image recognition and classification unit; the second image recognition and classification unit is used for acquiring the quantity and class information of the products in the product appearance image by utilizing the product recognition and classification model, and sending the quantity and class information of the products in the product appearance image at the current outlet to the database management module.
Further, the database management module comprises an inlet information receiving unit, an outlet information receiving unit and a data updating unit; the inlet information receiving unit is used for receiving the quantity and category information of the products at the inlet and sending the quantity and category information to the data updating unit; the export information receiving unit is used for receiving the quantity and category information of the products at the current export and sending the quantity and category information to the data updating unit; the data updating unit is used for updating the real-time inventory quantity of various products according to the quantity and the category information of the products at the current inlet/outlet.
Compared with the prior art, the invention has the following advantages: the neural network-based image recognition and classification method can be used for recognizing and classifying products in the product appearance image at the single outlet/inlet by utilizing the trained neural network model, further obtaining the product quantity and class information in the product appearance image at the single outlet/inlet, conveniently updating the real-time inventory quantity of various products based on the product quantity and class information, and avoiding the need of pre-manufacturing two-dimensional codes/bar codes, is relatively simple in operation, and reduces the cost of product in-and-out warehouse management to a certain extent.
Drawings
FIG. 1 is a schematic diagram of a neural network-based image recognition and classification system in accordance with an embodiment of the present invention;
fig. 2 is a flow chart of a neural network-based image recognition and classification method in an embodiment of the invention.
Detailed Description
The following describes in detail the examples of the present invention, which are implemented on the premise of the technical solution of the present invention, and detailed embodiments and specific operation procedures are given, but the scope of protection of the present invention is not limited to the following examples.
As shown in fig. 1, this embodiment provides a technical solution: the image recognition and classification system based on the neural network comprises a model training module, a warehouse shooting recognition module and a database management module;
in this embodiment, the model training module is configured to train the neural network through the sample image to which the class label has been added, so as to obtain a product identification classification model;
as a further step, the training process in the model training module is as follows:
s11: adding category labels to sample images of various products in a manual labeling mode to obtain a plurality of sample images with category labels, and forming a data set;
s12: dividing a sample image in a data set into a training set and a testing set according to the proportion of 3:7;
s13: inputting the sample images in the training set into a selected neural network, and training the network to obtain a trained network model;
s14: and (3) inputting the sample images in the test set into the trained network model in the step (S13), and when the performance index of the trained network model reaches a set value, storing parameters of the network model to obtain the product identification classification model.
As a further example, in step S13, the selected neural network includes, but is not limited to, an SSD network, a YoloV3 network, and the like.
When the product identification and classification model identifies the product appearance image, the product can be identified and the product target detection frame can be classified.
In this embodiment, the warehousing shooting identification module is configured to receive a warehousing shooting instruction sent by the upper computer, perform single shooting on a plurality of products at an entrance by using a first industrial camera, obtain a product appearance image, obtain the number and category information of the products in the product appearance image through a product identification classification model, and send the number and category information of the products in the product appearance image at the entrance to the database management module.
As a further step, the warehousing shooting identification module comprises a warehousing shooting instruction receiving unit, a first product shooting unit and a first image identification and classification unit; the storage shooting instruction receiving unit is used for receiving a storage shooting instruction sent by the upper computer and sending the instruction to the first product shooting unit; the first product shooting unit is used for shooting a plurality of products at the entrance by using a first industrial camera at the entrance according to the warehousing shooting instruction, obtaining a product appearance image, and sending the product appearance image to the first image identification and classification unit; the first image recognition and classification unit is used for acquiring the quantity and class information of the products in the product appearance image by utilizing the product recognition and classification model, and sending the quantity and class information of the products in the product appearance image at the current entrance to the database management module.
As a further step, the specific processing procedure of the first image recognition classification unit is as follows:
s21: acquiring a product appearance image at the current entrance, and classifying products in the image through a product identification classification model;
s22: counting the quantity information and the category information of the target detection frames of the products, namely acquiring the quantity and the category information of the products;
s23: and sending the quantity and the category information of the products at the current entrance to a database management module.
In this embodiment, the ex-warehouse shooting identification module is configured to receive an ex-warehouse shooting instruction sent by the host computer, perform single shooting on a plurality of products at an exit by using a second industrial camera, obtain a product appearance image, then obtain the number and category information of the products in the product appearance image through a product identification classification model, and send the number and category information of the products in the product appearance image at the exit to the database management module.
As a further step, the ex-warehouse shooting identification module comprises an ex-warehouse shooting instruction receiving unit, a second product shooting unit and a second image identification and classification unit; the ex-warehouse shooting instruction receiving unit is used for receiving an ex-warehouse shooting instruction sent by the upper computer and sending the instruction to the second product shooting unit; the second product shooting unit is used for shooting a plurality of products at the outlet by using a second industrial camera at the outlet according to the ex-warehouse shooting instruction, obtaining product appearance images and sending the product appearance images to the second image recognition and classification unit; the second image recognition and classification unit is used for acquiring the quantity and class information of the products in the product appearance image by utilizing the product recognition and classification model, and sending the quantity and class information of the products in the product appearance image at the current outlet to the database management module.
As a further step, the specific processing procedure of the second image recognition classification unit is as follows:
s31: acquiring a product appearance image at the current outlet, and classifying products in the image through a product identification classification model;
s32: counting the quantity information and the category information of the target detection frames of the products, namely acquiring the quantity and the category information of the products;
s33: and sending the quantity and category information of the products at the current outlet to a database management module.
In this embodiment, the database management module is configured to receive the number and the category information of the products at the single entry/exit, and update the real-time inventory number of each type of product according to the number and the category information of the products at the single entry/exit.
As a further feature, the database management module includes an entry information receiving unit, an exit information receiving unit, and a data updating unit; the inlet information receiving unit is used for receiving the quantity and category information of the products at the inlet and sending the quantity and category information to the data updating unit; the export information receiving unit is used for receiving the quantity and category information of the products at the current export and sending the quantity and category information to the data updating unit; the data updating unit is used for updating the real-time inventory quantity of various products according to the quantity and the category information of the products at the current inlet/outlet.
In this embodiment, the product appearance images are all products wearing the packaging box, and the network model realizes the identification and classification of the products according to the size of the packaging box.
As shown in fig. 2, the embodiment also provides an image recognition and classification method based on a neural network, which uses the system to realize the warehouse-in and warehouse-out management of products, and comprises the following steps:
s1: model training
Training the neural network through the sample image added with the class label to obtain a product identification classification model;
s2: warehouse entry management
Receiving a warehousing shooting instruction sent by an upper computer, shooting a plurality of products at an entrance by using a first industrial camera for one time to obtain a product appearance image, then obtaining the quantity and type information of the products in the product appearance image through a product identification classification model, and updating the real-time inventory quantity of each type of products according to the quantity and type information of the products in the product appearance image at the entrance;
s3: exit management
And receiving an ex-warehouse shooting instruction sent by the upper computer, shooting a plurality of products at the outlet by using a second industrial camera for a single time to obtain a product appearance image, then obtaining the quantity and type information of the products in the product appearance image through a product identification classification model, and updating the real-time inventory quantity of each type of product according to the quantity and type information of the products in the product appearance image at the outlet.
In summary, according to the neural network-based image recognition and classification method in the embodiment, the trained neural network model can be used to recognize and classify the products in the product appearance image at the single exit/entrance, so as to obtain the product quantity and category information in the product appearance image at the single exit/entrance, and based on the product quantity and category information, the real-time inventory quantity of various products can be conveniently updated, and two-dimensional codes/bar codes do not need to be prefabricated.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present invention, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
While embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the invention, and that variations, modifications, alternatives and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the invention.

Claims (9)

1. The image recognition and classification method based on the neural network is characterized by comprising the following steps of:
s1: model training
Training the neural network through the sample image added with the class label to obtain a product identification classification model;
s2: warehouse entry management
Receiving a warehousing shooting instruction sent by an upper computer, shooting a plurality of products at an entrance by using a first industrial camera for one time to obtain a product appearance image, then obtaining the quantity and type information of the products in the product appearance image through a product identification classification model, and updating the real-time inventory quantity of each type of products according to the quantity and type information of the products in the product appearance image at the entrance;
s3: exit management
And receiving an ex-warehouse shooting instruction sent by the upper computer, shooting a plurality of products at the outlet by using a second industrial camera for a single time to obtain a product appearance image, then obtaining the quantity and type information of the products in the product appearance image through a product identification classification model, and updating the real-time inventory quantity of each type of product according to the quantity and type information of the products in the product appearance image at the outlet.
2. The neural network-based image recognition and classification method of claim 1, wherein: in the step S1, the training process of the product identification classification model is as follows:
s11: adding category labels to sample images of various products in a manual labeling mode to obtain a plurality of sample images with category labels, and forming a data set;
s12: dividing a sample image in a data set into a training set and a testing set according to the proportion of 3:7;
s13: inputting the sample images in the training set into a selected neural network, and training the network to obtain a trained network model;
s14: and (3) inputting the sample images in the test set into the trained network model in the step (S13), and when the performance index of the trained network model reaches a set value, storing parameters of the network model to obtain the product identification classification model.
3. The neural network-based image recognition and classification method of claim 2, wherein: in the step S13, the neural network includes an SSD network and a YoloV3 network.
4. The neural network-based image recognition and classification method of claim 1, wherein: in the step S2, the specific process of acquiring the number and class information of the products in the product appearance image through the product identification classification model is as follows:
s21: acquiring a product appearance image at the current entrance, and classifying products in the image through a product identification classification model;
s22: and counting the number information and the category information of the target detection frames of the product, namely acquiring the number and the category information of the product.
5. The neural network-based image recognition and classification method of claim 1, wherein: in the step S3, the specific process of acquiring the number and the category information of the products in the product appearance image through the product identification classification model is as follows:
s31: acquiring a product appearance image at the current outlet, and classifying products in the image through a product identification classification model;
s32: and counting the number information and the category information of the target detection frames of the product, namely acquiring the number and the category information of the product.
6. The neural network-based image recognition and classification method of claim 5, wherein: the image recognition and classification method based on the neural network utilizes an image recognition and classification system based on the neural network to realize the warehouse-in and warehouse-out management work of products, and the image recognition and classification system based on the neural network comprises a model training module, a warehouse-in shooting recognition module, a warehouse-out shooting recognition module and a database management module;
the model training module is used for training the neural network through the sample image added with the class label to obtain a product identification classification model;
the system comprises a database management module, a product appearance image acquisition module, a product identification and classification module, a product appearance image acquisition module and a database management module, wherein the database management module is used for acquiring a product appearance image of a plurality of products at an entrance by utilizing a first industrial camera, acquiring the quantity and class information of the products in the product appearance image by utilizing a product identification and classification model, and transmitting the quantity and class information of the products in the product appearance image at the entrance;
the ex-warehouse shooting identification module is used for receiving an ex-warehouse shooting instruction sent by the upper computer, shooting a plurality of products at the outlet by using the second industrial camera for a single time to obtain a product appearance image, then obtaining the quantity and type information of the products in the product appearance image through the product identification classification model, and sending the quantity and type information of the products in the product appearance image at the outlet to the database management module;
the database management module is used for receiving the quantity and the category information of the products at the single entry/exit, and updating the real-time inventory quantity of various products according to the quantity and the category information of the products at the single entry/exit.
7. The neural network-based image recognition and classification method of claim 6, wherein: the warehousing shooting identification module comprises a warehousing shooting instruction receiving unit, a first product shooting unit and a first image identification and classification unit; the storage shooting instruction receiving unit is used for receiving a storage shooting instruction sent by the upper computer and sending the instruction to the first product shooting unit; the first product shooting unit is used for shooting a plurality of products at the entrance by using a first industrial camera at the entrance according to the warehousing shooting instruction, obtaining a product appearance image, and sending the product appearance image to the first image identification and classification unit; the first image recognition and classification unit is used for acquiring the quantity and class information of the products in the product appearance image by utilizing the product recognition and classification model, and sending the quantity and class information of the products in the product appearance image at the current entrance to the database management module.
8. The neural network-based image recognition and classification method of claim 7, wherein: the ex-warehouse shooting identification module comprises an ex-warehouse shooting instruction receiving unit, a second product shooting unit and a second image identification and classification unit; the ex-warehouse shooting instruction receiving unit is used for receiving an ex-warehouse shooting instruction sent by the upper computer and sending the instruction to the second product shooting unit; the second product shooting unit is used for shooting a plurality of products at the outlet by using a second industrial camera at the outlet according to the ex-warehouse shooting instruction, obtaining product appearance images and sending the product appearance images to the second image recognition and classification unit; the second image recognition and classification unit is used for acquiring the quantity and class information of the products in the product appearance image by utilizing the product recognition and classification model, and sending the quantity and class information of the products in the product appearance image at the current outlet to the database management module.
9. The neural network-based image recognition and classification method of claim 8, wherein: the database management module comprises an inlet information receiving unit, an outlet information receiving unit and a data updating unit; the inlet information receiving unit is used for receiving the quantity and category information of the products at the inlet and sending the quantity and category information to the data updating unit; the export information receiving unit is used for receiving the quantity and category information of the products at the current export and sending the quantity and category information to the data updating unit; the data updating unit is used for updating the real-time inventory quantity of various products according to the quantity and the category information of the products at the current inlet/outlet.
CN202311340461.9A 2023-10-17 2023-10-17 Image recognition and classification method based on neural network Pending CN117315557A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311340461.9A CN117315557A (en) 2023-10-17 2023-10-17 Image recognition and classification method based on neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311340461.9A CN117315557A (en) 2023-10-17 2023-10-17 Image recognition and classification method based on neural network

Publications (1)

Publication Number Publication Date
CN117315557A true CN117315557A (en) 2023-12-29

Family

ID=89297020

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311340461.9A Pending CN117315557A (en) 2023-10-17 2023-10-17 Image recognition and classification method based on neural network

Country Status (1)

Country Link
CN (1) CN117315557A (en)

Similar Documents

Publication Publication Date Title
CN108416403B (en) Method, system, equipment and storage medium for automatically associating commodity with label
CN111539668B (en) Warehousing system based on radio frequency technology and control method
US20230177391A1 (en) Machine-learning data handling
CN110414638A (en) A kind of stereo warehouse management system and management method based on RFID
CN107977815A (en) Warehouse management system and method
CN107045641A (en) A kind of identification of pallets method based on image recognition technology
CN102119343B (en) Target detection device and target detection method
CN111222389B (en) Analysis method and system for commodity on super commodity shelf
KR101059801B1 (en) Automatic logistics system and control method
CN106779565A (en) Logistic storage information management system based on intelligent monitoring
CN103810604A (en) Commodity price adjustment management tracking system and commodity price adjustment management tracking method
WO2021179138A1 (en) Method and system for analyzing commodities on supermarket shelf
CN112001669A (en) Engineering material control method, system and computer storage medium
CN116050992A (en) Inventory checking management system based on image recognition
CN116409575A (en) Warehouse-in and warehouse-out management and control system of intelligent storage shelf
CN117315557A (en) Image recognition and classification method based on neural network
CN110175564A (en) Client based on recognition of face revisits identification system and discrimination method
CN110309737A (en) A kind of information processing method applied to cigarette sales counter, apparatus and system
CN111222941B (en) Online purchase information classification system based on big data and handheld operation terminal
CN110163311A (en) A kind of logistic management system based on mobile Internet
Stojkic et al. The enchanting of information systems with digital technologies
CN116403339B (en) Printer intelligent management system and method based on RFID tag identification
KR102634031B1 (en) Universal invoice reader for identification of the courier company and method and system for collecting cargo information using the reader
CN110188847B (en) Donkey-hide gelatin traceability method and system based on NFC
CN117314478B (en) RFID-based product full life cycle management and control system and method

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