Semantic Communication Method Based on Compression Ratio Optimization for Vision Tasks in the Artificial Intelligence of Things

With the increasing amount of image data required for vision tasks, traditional data transmission methods face the problems of high latency and low reliability, which greatly limit the performance and application potential of the Artificial Intelligence of Things (AIoT). In order to solve this problem, this paper proposes a semantic communication method based on compression ratio optimization, which aims to reduce the communication delay and improve the reliability of data transmission by optimizing the compression and transmission process of image data. First, in the feature extraction phase the method extracts key visual features from the original image data. Subsequently, in the semantic relation extraction phase, we further analyze the association relations between these features to understand the semantic content of the image data. Finally, during the semantic compression stage, we implemented an adaptive optimization method that automatically selects the optimal compression ratio. Compared to traditional methods of wireless data transmission, this approach can significantly minimize transmission delay while ensuring that no critical data is lost. The experimental results show that this semantic communication method not only significantly reduces the completion time of the intelligent task, but also improves the antijamming ability of the image semantic communication.