1996 Fiscal Year Final Research Report Summary
Roles of street scene impression in navigation in city spaces
| Project/Area Number |
07680398
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Kyoto Institute of Technology |
|---|
Principal Investigator |
KUROKAWA Takao Kyoto Institute of Technology, Faculty of Engineering and Design, Professor, 工芸学部, 教授 (00029539)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
OGATA Masahito Kyoto Institute of Technology, Faculty of Engineering and Design, Research Assoc, 工芸学部, 助手 (50273545)
|
|---|
| Project Period (FY) |
1995 – 1996
|
| Keywords | navigation / city space / information space / street scene impression / behavior of predestrians / support of navigation |
|---|
| Research Abstract |
Guiding information for navigating in city spaces was analyzed based on the observation of pedestrians and the experiments of place identification, and the role of scene impression and spatial structure of streets was considered. 1. We observed the behavior of pedestrians along simple or complex paths and analyzed what kinds of information guided them to their destination. At crossings they often got keys of correct routes from scene impression along each streets. 2. Images of street scenes from which scene information was omitted to a various degree were presented to subjects and they were asked to identify the presented places. They responded most correctly for the scenes with a perspective angle. The subjects identified places on the basis of the structure of viewing spaces. It was confirmed that they failed frequently to identify as the amount of omitted information about key buildings was increased. 3. To the subjects we presented images of street scenes from an advancing point of vision and asked them to identify presented places. The results show that spatial arrangement of buildings and their color were used for place identification. The subjects took much time to respond and their error rate was increased when the whole or a part of information on street scenes was omitted. Although there were more correct responses for moving images than for static images, there was no difference in response time between the two. 4. We required subjects to do navigation tasks in a virtual city space constructed on a computer and recorded their behavior. This experiment revealed that people took one of three strategies for navigating in the virtual space. 5. In order to support navigation on Yedo maps drawn in the Yedo era we proposed a layred structure of a Yedo map database in which each original Yedo map was remapped onto a map with no distortion in direction and scale.
|
