CN-121985367-A - Commodity identification method and device based on Wi-Fi MIMO signals, electronic equipment and medium

Abstract

The application discloses a commodity identification method and device based on Wi-Fi MIMO signals, electronic equipment and a medium, and belongs to the technical field of data processing. The method comprises the steps of obtaining multiple pieces of CSI channel information acquired for a target commodity based on Wi-Fi MIMO signals, respectively extracting CSI characteristic information for the multiple pieces of CSI channel information, fusing the multiple pieces of CSI characteristic information to obtain fused characteristic data, and identifying the target commodity based on the fused characteristic data to obtain an identification result. According to the embodiment of the application, cloud intervention is not needed in the commodity identification process, so that processing lag caused by network transmission delay, cloud computing fluctuation and large video data volume is avoided, and a scene with no stability of the network can be dealt with.

Inventors

• SUN GUANGCHENG

Assignees

• 珠海艾维普信息技术有限公司
• 珠海格力电器股份有限公司

Dates

Publication Date

20260505

Application Date

20251230

Claims (10)

1. 1. The commodity identification method based on Wi-Fi MIMO signals is characterized by comprising the following steps: acquiring various CSI channel information acquired for a target commodity based on Wi-Fi MIMO signals; respectively extracting CSI characteristic information for the plurality of CSI channel information; fusing the plurality of CSI characteristic information to obtain fused characteristic data; And identifying the target commodity based on the fusion characteristic data to obtain an identification result.
2. 2. The method of claim 1, wherein the identifying the target commodity based on the fused feature data, resulting in an identification result, comprises: acquiring fingerprint characteristic data of one or more commodities in a preset fingerprint library; And identifying the target commodity based on the fusion characteristic data and the fingerprint characteristic data to obtain an identification result.
3. 3. The method according to claim 2, wherein the identifying the target commodity based on the fusion feature data and the fingerprint feature data, to obtain an identification result, includes: Determining a first feature vector corresponding to the fusion feature data; Determining a second feature vector corresponding to the fingerprint feature data; Determining vector distance data for the first vector data and the second vector data; and identifying the target commodity based on the vector distance data to obtain an identification result.
4. 4. A method according to claim 3, wherein said determining vector distance data for said first vector data and said second vector data comprises: determining euclidean distance data for the first vector data and the second vector data; Or, determining cosine similarity of the first vector data and the second vector data.
5. 5. The method according to any one of claims 1 to 4, wherein identifying the target commodity based on the fusion characteristic data to obtain an identification result includes: Judging whether the target commodity can be identified or not based on the fusion characteristic data; And outputting a commodity ID corresponding to the target commodity when the target commodity can be identified.
6. 6. The method as recited in claim 5, further comprising: And when the target commodity can not be identified, inputting the CSI characteristic information into a preset deep learning classification model, and outputting an identification abnormality reason.
7. 7. The method of claim 1, wherein the CSI channel information comprises any one or more of amplitude information, phase information, multipath propagation information.
8. 8. A commodity identification device based on Wi-Fi MIMO signals, the device comprising: The CSI channel information acquisition module is used for acquiring various CSI channel information acquired for the target commodity based on Wi-Fi MIMO signals; the CSI characteristic information extraction module is used for respectively extracting CSI characteristic information aiming at the plurality of CSI channel information; The fusion characteristic data determining module is used for fusing the plurality of CSI characteristic information to obtain fusion characteristic data; And the commodity identification module is used for identifying the target commodity based on the fusion characteristic data to obtain an identification result.
9. 9. An electronic device comprising a processor, a memory, and a program or instruction stored on the memory and executable on the processor, which when executed by the processor, implements the steps of the Wi-Fi MIMO signal-based commodity identification method according to any one of claims 1 to 7.
10. 10. A readable storage medium, wherein a program or instructions is stored on the readable storage medium, which when executed by a processor, implements the steps of the Wi-Fi MIMO signal-based article identification method of any of claims 1-7.

Description

Commodity identification method and device based on Wi-Fi MIMO signals, electronic equipment and medium Technical Field The application belongs to the technical field of data processing, and particularly relates to a commodity identification method based on Wi-Fi MIMO signals, a commodity identification device based on Wi-Fi MIMO signals, electronic equipment and a readable storage medium. Background At present, a commodity identification method is that a video is uploaded to a cloud end, a server carries out commodity identification on the cloud end video, but the commodity identification is realized by intensively processing video streams through the cloud end server, and the real-time identification is poor and cannot adapt to a network unstable scene due to network transmission delay, cloud end calculation fluctuation and processing lag caused by large video data quantity. Therefore, it is difficult to meet the demands of the unattended vending cabinet for low latency and high reliability. Disclosure of Invention The embodiment of the application aims to provide a commodity identification method based on Wi-Fi MIMO signals, which can solve the problems of network transmission delay, cloud computing power fluctuation and processing delay caused by large video data volume in current commodity identification. In order to solve the technical problems, the application is realized as follows: In a first aspect, an embodiment of the present application provides a commodity identification method based on Wi-Fi MIMO signals, where the method includes: acquiring various CSI channel information acquired for a target commodity based on Wi-Fi MIMO signals; respectively extracting CSI characteristic information for the plurality of CSI channel information; fusing the plurality of CSI characteristic information to obtain fused characteristic data; And identifying the target commodity based on the fusion characteristic data to obtain an identification result. Optionally, the identifying the target commodity based on the fusion characteristic data to obtain an identification result includes: acquiring fingerprint characteristic data of one or more commodities in a preset fingerprint library; And identifying the target commodity based on the fusion characteristic data and the fingerprint characteristic data to obtain an identification result. Optionally, the identifying the target commodity based on the fusion feature data and the fingerprint feature data to obtain an identification result includes: Determining a first feature vector corresponding to the fusion feature data; Determining a second feature vector corresponding to the fingerprint feature data; Determining vector distance data for the first vector data and the second vector data; and identifying the target commodity based on the vector distance data to obtain an identification result. Optionally, the determining the vector distance data of the first vector data and the second vector data includes: determining euclidean distance data for the first vector data and the second vector data; Or, determining cosine similarity of the first vector data and the second vector data. Optionally, identifying the target commodity based on the fusion characteristic data to obtain an identification result, including: Judging whether the target commodity can be identified or not based on the fusion characteristic data; And outputting a commodity ID corresponding to the target commodity when the target commodity can be identified. Optionally, the method further comprises: And when the target commodity can not be identified, inputting the CSI characteristic information into a preset deep learning classification model, and outputting an identification abnormality reason. Optionally, the CSI channel information includes amplitude information, phase information, and multipath propagation information. In a second aspect, an embodiment of the present application provides an apparatus for commodity identification based on Wi-Fi MIMO signals, where the apparatus includes: The CSI channel information acquisition module is used for acquiring various CSI channel information acquired for the target commodity based on Wi-Fi MIMO signals; the CSI characteristic information extraction module is used for respectively extracting CSI characteristic information aiming at the plurality of CSI channel information; The fusion characteristic data determining module is used for fusing the plurality of CSI characteristic information to obtain fusion characteristic data; And the commodity identification module is used for identifying the target commodity based on the fusion characteristic data to obtain an identification result. In one embodiment of the present application, the article identification module 404 may include: the fingerprint feature data acquisition sub-module is used for acquiring fingerprint feature data of one or more commodities in a preset fingerprint library; And the identification