Face expression image detection and recognition based on big data technology

This research addresses the deficiencies in current dynamic sequence facial expression recognition methods, which suffer from limited accuracy and effectiveness. The primary objective is to introduce an innovative approach that leverages big data technology for improved facial expression detection and recognition. The methodology encompasses several vital steps. The integral graph method is employed to capture dynamic sequences of facial expressions, and a weak facial feature classifier is utilized for image preprocessing. To enhance accuracy, a dynamic sequence model is devised for feature extraction. The study combines the personalized learning algorithm with the optical flow technique to pinpoint critical facial expression junctures and facilitate dynamic sequence recognition. The investigation reveals the inadequacy of current dynamic sequence facial expression recognition methods in accurately categorizing expressions. The proposed approach yields promising results, achieving a peak expression division accuracy of 91.78% in simulations. Notably, the personalized learning recognition method demonstrates enhanced robustness in categorizing expressions, effectively capturing intricate facial details and augmenting overall recognition efficacy. This research thus contributes to advancing facial expression recognition technology, addressing critical shortcomings in current methods.