| Budget Amount *help |
¥4,100,000 (Direct Cost: ¥4,100,000)
Fiscal Year 2002: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 2001: ¥800,000 (Direct Cost: ¥800,000)
Fiscal Year 2000: ¥2,400,000 (Direct Cost: ¥2,400,000)
|
|---|
| Research Abstract |
Although facial information such as face and facial expressions seems to be important in term of recognition of emotion and/or KANSEI and also communication between human and machine, there are no general-purpose software developed for facial information because of difficulty in image processing. We therefore develop the software which can extract outline of facial organs such as eyebrows, eyes, mouth and nose stably and precisely, on Microsoft Windows 98. For the purpose of good usability with respect to application software development, we provide software functions for automatic extraction of facial image date into software is also important for our software, we prepare API for image acquisition and format transformation
In order to investigate the important factor for recognition of facial expressions, we focus on extract differences between healthy people and schizophrenic people who has specific feature in terms of facial expressions. It is also very important because of necessity and demand for quantitative information for the diagnosis of schizophrenic. We have developed 1. an automatic interview system, 2. a method for the automatic extraction of facial organs, and 3. A method for the acquisition of the quantitative information for diagnosis, and tried to obtain quantitative data for diagnosis by analyzing facial expressions during the automatic interviews. In this study, we focus on the movement of the pupils and head the correlation between them, i.e. we develop a method for the automatic measurement of time sequential movements with respect to 1. the position of the pupil relation to the right and left corners of the eye, and 2. The direction of the frontal face. By calculating the correlation between them, we obtain the quantitative information which enables us to diagnose whether a subject is schizophrenic or not
|
