http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Luminance Correction for Image Projection on a Checked-Pattern Screen
Makiko Sembon,Mie Sato,Naoki Hashimoto,Masao Kasuga 대한전자공학회 2007 ITC-CSCC :International Technical Conference on Ci Vol.2007 No.7
We propose a method of the luminance correction for the image projection on a checked-pattern screen. Subjective evaluation shows that this technique can project images onto a patterned planar surface without giving annoyance with the pattern to viewers. It is applicable to home theater, exterior advertisement and so on.
Consideration of a Face Recognition System with Real-time and Robustness
Hitoshi Horiuchi,Satoru Kaneko,Mie Sato,Koichi Ozaki,Masao Kasuga 대한전자공학회 2007 ITC-CSCC :International Technical Conference on Ci Vol.2007 No.7
Capability of recognition, collision avoidance and communication with humans in natural is required for robots coexisting with humans. It is important that real-time and robust image processing for face recognition in a real environment. In this paper, we have developed a method of face recognition no depending on direction and size of human faces on various images. This method is include preprocess of a pattern matching algorithm based on the HSV color coordinate system and a rapid search algorithm by using the Genetic Algorithm (GA). For accurate face recognition, extracted images by an edge filter are used when a value of fitness on GA is evaluated. As a result, this system extracted face with about 65% accuracy.
Extracting Human Features Based on Fractal Dimension of Human Skin Surface Images
Ali Parchamy Araghy,Mie Sato,Masao Kasuga 대한전자공학회 2007 ITC-CSCC :International Technical Conference on Ci Vol.2007 No.7
In this paper, we measure fractal dimension of human skin surface images taken with an USB-microscope camera based on the power spectrum method and applied it to feature extraction of human skin. The results indicated that the human surface image could be represented by fractal dimension and this approach is effective and reliable to feature extraction from skin surface images.