Published Paper Details:

An enhanced method has been proposed in this work to detect different face angles of the same individual and the ability to recognize them correctly within appreciable amount of time is being presented. Image enhancement is an important part of facial recognition. In this work, Histogram Equalization is employed to adjust the image intensity and enhance the contrast. Feature extraction is a significant step to detect and subsequently recognize the face. To extract more productive features from these static images, this work recommends an algorithm based on the concatenation of the Modified Pixel Difference Value (MPDV) and Two Power Term features. MPDV is a modified version of PDV where due to immense pixel preservation the information in the image is highly conserved. Since the Two Power Term follows a pattern that is uniformly growing it increases the information of the pixels by a great amount that helps in obtaining efficient features to be compared. To generate a row vector, the features are concatenated from above two techniques, which are restructured into a column vector. Utilizing Euclidean distance column vectors of all the images from the database are compared against the column vectors of the test image. To perceive the individual minimum distance between the specific image and the location of the image in the database is obtained.