The difficulty of criminal psychological recognition is that it is difficult to classify emotions, and the accuracy of traditional recognition methods is insufficient. Therefore, it is necessary to improve the accuracy rate in combination with modern computer technology. This study uses deep learning as technical support and combines EEG computer signals to classify criminal psychological emotions. Moreover, a method for classifying EEG signals based on the state of mind of neural networks was constructed in the study. In addition, the EEG is denoised preprocessed by time-domain regression method, and features of the EEG signal parameters of different criminal psychological tasks are extracted and used as the input of the neural network. Finally, in order to verify the effectiveness of the algorithm, a simulation experiment is designed to study the effectiveness of the algorithm. The results show that the method proposed in this paper has certain practical effects.

犯罪心理识别的难点在于难以对情感进行分类，传统识别方法的准确性还不够。因此，有必要结合现代计算机技术来提高准确率。本研究使用深度学习作为技术支持，并结合脑电图计算机信号对犯罪心理情绪进行分类。此外，本研究还建立了一种基于神经网络心态的脑电信号分类方法。另外，通过时域回归方法对脑电信号进行去噪预处理，提取不同犯罪心理任务的脑电信号参数特征​​，作为神经网络的输入。最后，为了验证算法的有效性，设计了一个仿真实验来研究算法的有效性。结果表明，本文提出的方法具有一定的实际效果。