Identification and Decoding Technology of EEG Under Complex Visual Stimuli Using Clustering Ensemble

Accurate target recognition in complex backgrounds is the primary key step for aircraft to achieve sensitive perception and control. Traditional computer vision based object recognition technology is difficult to fully meet practical application needs for complex and open flight environment scenarios such as camouflage and variants. Its recognition accuracy and other performance are still far lower than manual interpretation. This article innovatively proposes an analysis method for identifying and decoding EEG signals under complex visual stimuli based on clustering ensemble. Firstly, this paper design the method for collecting and preprocessing EEG under visual stimulation obtain effective EEG information that can be represented in relevant visual stimuli; Further research was conducted on the time-domain and time-frequency feature extraction techniques for EEG signals to obtain EEG feature information with effective discriminative power under different stimulation; Finally, a clustering model for multi domain and multi-scale EEG mixed features was constructed, and a clustering ensemble model construction method for multiple subjects was designed to achieve strong generalization EEG identification and decoding across subjects under visual stimuli, thereby realizing perceptual recognition of relevant visual stimuli. The experimental results in this paper indicate that the proposed approach can precisely identify the target of interest in the image, promote the quantitative analysis of human brain signal representation models under visual stimuli, and promote the development of new brain inspired target recognition technologies.