New Born Face Feature Extraction Recognition using Image Processing Techniques

Abstract

Biometric techniques such as fingerprints, palm prints, irises, faces, and retinas are used to identify the individuals responsible for the forgery. The importance of facial recognition has grown as a direct result of the rise in the number of fakes. The information that makes up the repository for newborn babies comes from a wide variety of sources, and it features babies in a variety of poses and lighting conditions throughout many distinct galleries. A number of different facial recognition strategies and classification algorithms currently used in commercial systems are also examined as part of a standard benchmark test. The data set is obtained from the face of a newborn baby, then trained and tested with classification algorithms, and then compared based on a variety of performance indicators. The extractable features from a picture serve as the basis for the classification process, and some of these extracted features are also used for the purposes of training and testing in conjunction with the classification process. Different feature extraction strategies are investigated in this study, including local binary patterns (LBP), principal component analysis (PCA), and gray level co-occurrence matrix (GLCM). LBPs are described by their local binary patterns. Eigenfaces and Eigenvectors are produced using Principal Component Analysis, and second-order statistical features are constructed with Gray Level Co-occurrence Matrix. Photographs of newborn babies displaying a variety of facial expressions use these techniques. The extracted features are provided as input to the support vector machine for classification. In comparison to the other feature extraction methods, the principal component analysis method had an accuracy of 91%, a better recognition rate, and a shorter computation time. Experiments have also shown that this method is superior to other.