Proposing a Novel Three Triangles Geometrical Approach for Face Detection on Occluded Facial Features

Bikash Lamsal, Noriko Kojima, Naofumi Matsumoto
2016 Journal of the Institute of Industrial Applications Engineers  
This paper is associated with the novel approach for detecting the face and facial features from the images where the facial features are occluded. Face detection is one of the developing trends in the field of biometrics. A lot of face detection algorithm has been developed, but still the face detection rate haven't reached to 100%. It is due to the presence of various environmental conditions on an image. The occluded facial image is also one of the conditions for high performance face
more » ... on. A face is occluded if some area of the face is hidden behind on an object like a mask, sunglass, hand or we can say the face is partially visible. In this paper, we have developed a system, that detects the presence of the facial features and the face from the still images even the facial components are occluded. In this system, we calculate the geometrical position of the facial components by using the facial geometry by calculating the distance and the location of the facial features and the width of the face. The system first extracts the eyes, nose and mouth also known as the facial features by using the skin color model and the Haar cascades. The geometrical model is then applied to the extracted features using the three triangle method that calculates the height, width, distance and the position of the extracted features. Our proposed face detector then detects the presence of the face in an image even with an occlusion. We are developing the face detector for still film image by combining the "three triangle method" with "unscented Kalman filter (UKF)" and "Haar cascade classifier", which represents the novelty of our proposed system. We performed our experiments on various still film images under different conditions, using different facial databases to clarify the effectiveness of our proposed method.
doi:10.12792/jiiae.4.192 fatcat:o3xvp574d5go5abfciekrbml34