Human face detection in color images using skin color and template matching models for multimedia on the Web

Abstract

Human face detection techniques play an important role in applications like face recognition, video surveillance, human computer interface, face image database management, and querying image databases. Using color information in images is one of the various possible techniques usedfor face detection. This paper proposes a novel techniquefor detecting faces in color images using an adaptive threshold and template matching techniques. The goal of the technique is to segment the skin regions from the non-skin regions. Experimental results demonstrate successful face detection over a wide range offacial variations in color, position, scale, orientation, 3D pose, and expression in images from several photo collections (both indoors and outdoors). This method is quite practical andfaster when compared to neural networks and other techniques especially suited well for multimedia applications on the web.

Visual words for lip-reading

“…The work done by Nallaperumal et al (2006) fits into this category. Their method segments the image depending on the colour of the skin, using an adaptive threshold to separate regions of candidate faces from background.…”

Framework for Face Detection

“…The others are set to zero. After this, we iterate through each of the regions found in order to determine if the region might suggest a human face or not [1]. Now, for each of the connected regions or labeled regions we measure a set of properties like Euler no., eccentricity and bounding box.…”

“…To include the variations of the face sizes and also to consider the fact that the neck region is very likely to be part of the connected components if exposed, we take this ratio between 0.8 and 1.6. If the ratio of any region lies between this it is taken as face [1].…”

“…Face detection is a process that aims to locate a human face in an image. The task of face detection is so trivial for the human brain, yet it still remains a challenging and difficult problem to enable a computer to do face detection [1]. This is because human face is a dynamic object and has a high degree of variability in its appearance, which makes face detection a difficult problem.…”

Video based vomit and facial foreign object detections for baby watch and safety

“…Finally, they use these features as bases for facial condition recognitions and other applications. In the facial region extractions, they can be divided into the following techniques: Color analysis method [1,2,3,4], template matching method [5,6], feature method [7,8,9], and neural network method [10,11]. In the facial feature detection, they can also be divided into the following methods: Color analysis method [12], template matching method [13], and geometric feature method [14].…”