| Project/Area Number |
10555182
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B).
|
| Allocation Type | Single-year Grants |
|---|
| Section | 展開研究 |
|---|
| Research Field |
交通工学・国土計画
|
|---|
| Research Institution | Hokkaido University |
|---|
Principal Investigator |
HAGIWARA Toru Graduate School of Engineering, Hokkaido University, Assistant Professor., 大学院・工学研究科, 助教授 (60172839)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
NARISADA Kouhei School of Letters, Chukyo University, Professor., 文学部, 教授 (30237599)
|
|---|
| Project Period (FY) |
1998 – 2000
|
| Project Status |
Completed (Fiscal Year 2000)
|
| Budget Amount *help |
¥3,500,000 (Direct Cost: ¥3,500,000)
Fiscal Year 2000: ¥800,000 (Direct Cost: ¥800,000)
Fiscal Year 1999: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 1998: ¥1,800,000 (Direct Cost: ¥1,800,000)
|
|---|
| Keywords | Visibility / Digital still camera / Image processing / contrast / Charged coupled device / Traffic |
|---|
| Research Abstract |
In this research, three experiments were conducted to apply the digital image to improve the visibility of the road space. First experiment was to determine whether a digital still camera can be used to measure luminous contrast with the same degree of accuracy as that obtained by using a photometer. Second experiment was to compare the visibility performance of a digital still camera and the human eye. On the first experiment, luminance was measured during the daytime using a photometer and a digital still camera. The last experiment was to assess the visibility of road scene in fog in the daytime using the digital image. The digital still camera used in this study is one of the highest quality cameras currently available, a Canon EOS-DCS3. The photometer used was a Prichard PR-880. A gray scale was used as a reference target. The photometer and the digital still camera were placed in close proximity. The photometer directly measured luminance of the target, whereas the digital still
… More
camera took a picture of the target at the same time. The image obtained from a digital still camera can be analyzed in terms of the average and distribution of digitized brightness. Three or four combinations of aperture and shutter speed were used for each exposure value. A model for the conversion of digitized brightness into luminance was developed as a function of exposure value using the method of least squares. Thus, an image taken by the digital still camera can provide the luminance value as a function of exposure value. The second experiment was conducted on a test site for the driver's license in Sapporo. The test section is straight and flat, runs south to north, and is 20m wide and 300m long. Three targets (dark gray, medium gray and light gray) were set up in the center of the test section. The experimenter took two images using the digital still camera during the exposure period. The background luminance and contrast were estimated using the conversion model. The detection level decreased as the contrast decreased. In most cases, the low detection level fell well below Blackwell's threshold contrast of the human eye. Based on the results, it is concluded that a very high-quality digital still camera can be used to accurately evaluate visibility of targets on a road. The third experiment proposed basic methods of assessing poor visibility by using digital photo images. Within each images under adverse weather conditions, we found that there is little target's contrast, and that distribution of gray level is uniform. The power spectra of the combined distribution of gray level's frequencies decreased with increases in the fog density. We tried to classify the condition of fog in images by the each target's of contrast and the average of each power spectrum Less
|
