Deep learning system and method based on hyperspectral camera technology

A deep learning and hyperspectral technology, applied in the field of data processing, can solve the problems of low accuracy and low efficiency of unmanned supermarket monitoring and management, and achieve the effect of reasonable design and convenient use.

CN110009836AInactive Publication Date: 2019-07-12JIANGXI UNIV OF SCI & TECH
6 Cites 22 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
Publication Date
2019-07-12
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention provides a deep learning system and method based on the hyperspectral camera technology. The method comprises the following steps: storing face information of a user and identity information corresponding to the face information; shooting the face of a shopper by using a face recognition camera to obtain the face information of the shopper; comparing the face information of the shopper with the stored face information of the user, and when the face information of the shopper exists on a server, opening a shopping gate; shooting the action track of the shopper in real time, and uploading an action track image of the shopper to the server for storage; sending a changed weight value to the server, and reporting the position of a shelf; comparing the position of the shelf with the action track image of the shopper to obtain the shopper currently at the position of the shelf; obtaining unit prices of settlement items to be settled together; and when it is detected that a checkout request the shopper is completed, opening an exit door. By adoption of the deep learning system and method provided by the invention, accurate monitoring and management of shopping behaviors in unmanned supermarkets are achieved.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The present application relates to the technical field of data processing, in particular to a deep learning system and method based on hyperspectral imaging technology. Background technique

[0002] With the development of society and the advancement of science and technology, industrialized or intelligent industries are gradually replacing artificial industries. Whether in terms of industrial efficiency or the accuracy of control, management and production, industrialized or intelligent industries have their own unique advantages. The advantages. As an indispensable industrial place for modern people's life, supermarkets have naturally become one of the research focuses of this industrialized or intelligent industry. Unmanned supermarkets and supermarkets without salespersons are also called unmanned supermarkets. It is not the salesperson who is responsible for the cash register, but the It is a payment device, and the current unmanned supermarket i...

Examples

Embodiment

[0071] Such as figure 1 As shown, it is a schematic flow chart of the first method of deep learning based on hyperspectral imaging technology in this embodiment. Hyperspectral imaging is used to record shoppers' facial information and action trajectory images in unmanned supermarkets, combined with weight sensors on shelves Obtain the information of the items purchased by shoppers to complete the shopping information check and settlement check, which ensures the accuracy of the unmanned supermarket and reduces the management cost of the unmanned supermarket. The method comprises the steps of:

[0072] Step 101, storing the user's facial information and identity information corresponding to the facial information on the server.

[0073] Step 102, when it is detected that the shopper enters the unmanned supermarket, use the face recognition camera to shoot the shopper's face to obtain the shopper's facial information; upload the shopper's facial information to the server, compa...