Facial expression recognition method based on ROI-KNN convolutional neural network

The invention discloses a facial expression recognition method based on an ROI-KNN convolutional neural network. The method comprises the following steps: S1, acquiring video image data of a driver in a driving process; s2, performing face positioning on the video image data, and extracting a face feature region; s3, intercepting a video frame of the face feature region as face image data according to a preset time interval; s4, performing enhancement correction on the facial image data to obtain a facial expression image; s5, inputting the facial expression image into a trained deep convolutional neural network model; and S6, outputting a driver facial expression recognition result. Through the deep convolutional neural network model constructed in combination with the region of interest and the K-nearest neighbor algorithm, the training effect of the facial expression data in the recognition model can be improved, so that the problem of insufficient generalization ability of the deep convolutional neural ne