Design of a Parallel Algorithm for the Detection of Classic Facial Expressions

For a machine to recognize, model, and express human emotions, it needs a robust set of processes because it does not have the context to interpret verbal and body language. The physiological, visual, and vocal characteristics denote that emotions are natural and inherent in humans, not machines. This paper interprets facial expressions with an algorithm that serves as an intelligent human-computer interface; our approach uses video or images captured through real-time cameras. The literature shows widespread interest in machines detecting and interpreting human emotions. However, it is necessary to develop an intelligent interface for this to happen. Hence, the field of affective computing is derived. Affective computing studies how machines recognize, analyze, model, and represent emotions. Reeves and Nass showed that people treat computers, cell phones, and other electronic media as if they, too, were human beings with personalities, feelings, and even their own will. To this day, detecting facial expressions represents a problem to be solved; in the literature, many solutions have been proposed and approached differently. This article developed a parallel algorithm to detect expressions in real-time using the Haartraning algorithm. Facial movements are interpreted as active units; each allows the detection of the simple emotions of Robert Plutchik’s classification.