Expression Invariant Face Recognition System
Deepti Ahlawat, Vijay Nehra · International Journal of Signal Processing Image Processing and Pattern Recognition · 2017
Face recognition as one of the biometric technique is one of the interesting topic in the field of computer vision and pattern recognition.In the present study, attempt has been made in this direction to address the issues and challenges in face recognition system viz.pose, illumination, facial expressions etc .So, a method is presented which converts input face image with an arbitrary expression into its corresponding neutral facial expression image.In the present study, the method applies deep neural network to train the system to locate the landmarks in the face image.The landmark points create an intermediate triangular mesh and warp the image using affine transform.This warped image is classified using Support Vector Machine (SVM) and k-means classifier.Finally, results are compared and recognition accuracy is determined for different expressions.The result obtained from the investigated work is compared with state-of-art algorithms.For the proposed method the recognition rate is 97.3% for JAFFE database and 97.8% for CK database which outperforms the other state-of-art methods.The present work is tested on JAFFE and CK databases.