Local Directional Number Pattern for Face Analysis: Face and Expression Recognition
Technology Used: Dot Net
This paper proposes a novel local feature descriptor, local directional number pattern (LDN), for face analysis, i.e., face and expression recognition. LDN encodes the directionalÃƒâ€šÃ‚Â information of the faceÃƒÂ¢Ã¢â€šÂ¬Ã¢â€žÂ¢s textures (i.e., the textureÃƒÂ¢Ã¢â€šÂ¬Ã¢â€žÂ¢s structure) in a compact way, producing a more discriminative code than current methods. We compute the structure of each micro-pattern with the aid of a compass mask that extracts directional information,Ãƒâ€šÃ‚Â and we encode such information using the prominent direction indices (directional numbers) and signÃƒÂ¢Ã¢â€šÂ¬Ã¢â‚¬Âwhich allows us to distinguish among similar structural patterns that have different intensity transitions. We divide the face into several regions, and extract the distribution of the LDN features from them. Then, we concatenate these features into a feature vector, and we use it as a face descriptor. We perform several experiments in whichÃƒâ€šÃ‚Â our descriptor performs consistently under illumination, noise, expression, and time lapse variations. Moreover, we test our descriptor with different masks to analyze its performance in different face analysis tasks.