抄録
This study focuses on the development of method to posit view conditions of the forest videos (filmed every day by the "Robot Camera" in the forest of University of Tokyo located in Chichibu) with the help of meteorological data observed near the camera, and on the practicality verification of the method in case of applying it to collect the intended videos in foliage season. First, for the development of the method, view conditions are divided into 4 categories in terms of visibility and sunshine, by watching 3 shots by which view condition of that day is represented. Then, by the discriminant functions (by the logit model) of meteorological data, view conditions of each day are discriminated into the said 4 categories in error rate of about 0. 3. Second, the application of this method is tried on the assumption that users of these forest videos collect videos by which the typical sceneries in foliage season are filmed.In the result, it can be concluded that the method developed on this study is effectual in some purposes and that it is possible to enhance the practicality to consider several ways of application.
