Tahir Celik

Learn More
Facial feature extraction is a fundamental problem in image processing. Correct extraction of features is essential for the success of many applications. Typical feature extraction algorithms fail for low resolution images which do not contain sufficient facial detail. In this paper, a region-based super-resolution aided facial feature extraction method for(More)
Automated visual surveillance systems mostly depend on an effective background subtraction technique. Most background subtraction techniques suffer mainly from parameter updates for threshold selection. A new threshold selection technique, which is found while training the system to learn the background, is proposed.
Based on a combination of color segmentation, connected components labeling and morphology, an algorithm for human face tracking and facial feature based head tilt angle extraction is developed. The method uses the HSI space for color segmentation, since, in this space, the dynamic range of skin color is quite narrow, thus enabling the differentiation of(More)
Moving object edge detection and segmentation method is presented with utilizing multi-wavelets. The subsequent segmentation of moving objects is achieved by binary morphological operations. The proposed multi-wavelet based method is compared with the methods based on scalar wavelets using both single and double change detection techniques. The simulation(More)
In this paper, we propose a facial feature detection technique which uses combination of image intensity, color, and geometrical features of faces and provide a performance analysis of the technique proposed. The paper provides a cost factors analysis for optimizing the correct detection of facial features. In addition it also introduces the idea of(More)
Based on a combination of color segmentation, connected component labeling and morphology, an algorithm for human face tracking and principal component analysis (PCA) based head tilt angle calculation method is developed. The method uses HSV space for color segmentation since in this space the dynamic range of the skin color is quite narrow thus enabling us(More)
  • 1