Research on Face Image Recognition System Based on Computer Artificial Intelligence Technology

This paper develops a new facial recognition technology based on Python+ OpenCV technology. This study includes three modules: face recording, face training and face recognition. A sparse representation classification technique based on Gabor low - order restoration is presented to recognize face images. By Gabor transformation, the images of each training sample and their corresponding error images are transformed to obtain the feature vector of Gabor. The feature vectors of Gabor are combined into a dictionary of Gabor characteristics. The experimental results show that the algorithm can suppress the interference of variable light and preserve the facial expression well. This method can not only improve the brightness contrast of the image, but also filter out the light component of the face effectively. The accuracy of face recognition under different illumination conditions is 98.56%.