Multi-channel image semantic communication method and system, computer device
By employing a multi-channel image semantic communication method, utilizing a deep source encoder and a generative adversarial network model, the problem of communication instability in unreliable transmission environments is solved, achieving stable transmission and efficiency improvement of image data in unreliable transmission environments.
CN116800860BActive Publication Date: 2026-06-26NORTHEASTERN UNIV CHINA
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NORTHEASTERN UNIV CHINA
- Filing Date
- 2023-05-19
- Publication Date
- 2026-06-26
AI Technical Summary
Technical Problem
In unreliable transmission environments, when using unreliable transmission protocols for data transmission, the communication effect is unstable, which can easily lead to semantic ambiguity and affect communication quality.
Method used
A multi-channel image semantic communication method is adopted, which extracts image data features through a deep source encoder and uses multi-channel cooperative transmission and a generative adversarial network model for decoding, thereby achieving stable transmission of image data in unreliable transmission environments.
Benefits of technology
It improves the stability and efficiency of image data transmission, reduces transmission time, and enhances communication quality.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN116800860B_ABST
Abstract
The application discloses a kind of multi-channel image semantic communication method and system, computer equipment belong to computer communication technical field, mainly solve the unstable problem of communication effect when data transmission is carried out using unreliable transmission protocol in prior art, including information sending end obtains image data to be transmitted, and using depth source encoder is extracted to the feature of the image data to be transmitted Processed, obtain feature semantic information to be transmitted;The information sending end uses the mode of multi-channel cooperative transmission to the feature semantic information to be transmitted Parallel cooperative transmission is carried out to information receiving end, obtains multiple receiving end sub-information;The information receiving end carries out multi-channel feature fusion processing to the receiving end sub-information, obtains multi-channel fusion information;The information receiving end is decoded to the multi-channel fusion information based on generative adversarial network model Processing, obtains target image data.
Need to check novelty before this filing date? Find Prior Art