Facial Expression Synthesis Images Using Hybrid Neural Network With Particle Swarm Optimisation Techniques

In the advance life trend, the facial expression is the visual facial outer structure of the human affective state, intellectual action, and human interchanges and facial expression go about as the key role in the movement of communication. In this paper, the facial expression synthesis performance is done using different facial expressions such as angry, sad, smile, surprise and cry of various peoples. In the proposed method, two procedures are used namely hybrid neural network (HNN) and particle swarm optimisation (PSO) algorithm. By training particle swarm optimisation and hybrid neural network, we take the desired output. In the result section, various evaluation metrics namely peak signal to noise ratio (PSNR), mean square error (MSE) and a second-derivative-like measure of enhancement (SDME) value is calculated using diverse algorithms. In this evaluation performance, the particle swarm optimisation is given enhanced output while comparing it with other techniques and the existing methods of facial expression.