Published Paper Details:

The face recognition is a classical research domain, a number of research contributions are available by different contributors. Beyond their key contributions the proposed work is aimed to design a new recognition model that can work on incomplete data or information of face to recognize the target person. Additionally the work also indented to find pose recognition too. Both the issues are lead to increase the rate of misclassification or recognition during the face recognition. Thus the proposed work is proposing a LDA and MLP based face recognition to mitigate the issues of face recognition task. The proposed working model including the LDA based feature selection technique. The LDA (Linear Discriminant Analysis) is technique that is also used for dimensionality reduction additionally to compute a common projection in a higher dimensional space. Further the LDA based pose features are trained using the MLP (Multi Layer Perceptron). The MLP is a feed forward neural network and work for optimal classification outcomes. After training the model is tested on a test dataset. The experimental results demonstrate the effective performance in terms of accuracy as well as the method shows less resource consumption in terms of memory and time.