Skin segmentation based face tracking independent of lighting conditions
The field of image processing field is inundated with new ideas and concepts put forth by various researchers with an intention to overcome the challenges encountered during face detection. Among the various techniques proposed to achieve optimum face detection, skin segmentation has been most popular for its coherence and its minimalism in ascertaining faces from photos. However, the major obstacle while attempting to achieve exceptional face detection using skin segmentation is the disparity of the skin color when an image is captured under different lighting conditions. The hues of the captured skin vary from natural light to artificial indoor lighting to partial lighting and shadow effect of light. Also, considering the diverse skin textures of people worldwide, it is intricate to achieve immaculate skin detection in the presence of such disparity. The paper presents a novel technique designed to perform skin segmentation based face tracking independent of lighting conditions. The proposed skin segmentation algorithm has been implemented and tested to identify all possible skin tones under light variant conditions. Integration of the novel skin segmentation algorithm into a larger real-time face detection system yielded 99.9% accuracy and precision, thus achieving a flawless skin-segmentation based face tracking system.
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.