Detection of patterns in images using classifiers is one of the most promising topics of research in the field of computer vision. A large number of practical applications for face detection exist and contemporary work even suggests that any specialized detectors can be approximated by using fast detection classifiers. In this project, I have developed an algorithm which will detect face from the input image with less false detection rate using combined effects of computer vision concepts. This algorithm utilizes the concept of recognizing skin color, detecting edges and extracting different features from face. The result is supported by the statistics obtained from calculating the parameters defining the parts of the face. The project also implements the highly powerful concept of Support Vector Machine that is used for the classification of images into face and non-face class. This classification is based on the training data set and indicators of luminance value, chrominance value, saturation value, elliptical value and nose, eye & mouth map values.
Shah, Parin M., "Face Detection from Images Using Support Vector Machine" (2012). Master's Projects. 321.