|
|
|
Poster Session 3
|
The Study on generating of BRDF table set for Multi Channel Sensor
Asako Konda, Msatoshi Yokota, Taku Nishijima , Koji Kajiwara, Yoshiaki Honda
Center for Environmental Remote Sensisng (CERES), Chiba University
1-33. Yayoi-Cho , Inage-Ku Chiba 263-8522, Japan
Tel : (81)-43-290-3845 ; Fax : (81)-43-290-3857
E-mail: Konda@rsirc.cr.chiba-u.ac.jp
Abstract
We were measuring BRDF (Bi- Directional Reflectance Distribution Factor) for the purpose of multi channel sensor analysis. For that analysis we were measuring used tower (Height is 5m) and 2 stepping motor. In this time, tower was loading a truck, and we became to get a lot of data. In due to measure these study, we could possibly that understood characteristic of grassland BRDF. First step for making BRDF table set for grassland, we discussed Composite Algorithm that avoid effects of BRDF.
1. Introduction
Recently, satellite remote sensing is becoming important engineering for measurement of environmental. However , satellite data are not only information of object, but also including another information of atmosphere, geographic and etceteras. When we used satellite data and analyze, we should remove these informations.
BRDF (Bi- Directional Reflectance Distribution Factor) is in dependent on sun Zenith angle, sun ,sensor azimuth angle, geographic information, and that dependence are different to each sensor dependence of electronic wave. If we get the truth information, we should remove some other information. And BRDF is one of some another information. Usually, we use some MODEL due to express geometry dependence of multi reflectance. But these MODELS are no suitable for using correction and estimation algorithm because these are using some parameter including difficult measurement and calculation, such as distribution of leaves.
2. Objectives
Therefore, in this study, we the measured the ground truth for Mongolia grassland, and analyzed BRDF characteristic. And we development a BRDF table set for correcting satellite data.
3. Observation
3.1 Measurement System
We used measurement system that watanable et al., 1) developed at 1997. (Fig.1) Detail functions are shown the Table.1. And in this study, getting vegetation coverage system is change Digital video camera from 8mm video camera.
3.2 Observation
We carried out an observation the adobe system for the 23 rd July to 15 th August 1998 in Mongolia. Measurement areas are the three points with different vegetation coverage (Fig.2). Watanabe et al. were that one cycle of measurements every 30 degrees from 0 to 330 degrees for sensor azimuth angle and every 10 degrees from 0 degree to 60 degrees for sensor zenith angle. Total are 75 point measurements. But, by that measurement of large sensor zenith angle is less point than measurement of small sensor zenith angle. Therefore our measurement were following Table.2. Time of one cycle require about 15 minutes. It is more short time than them time because our control PC is more powerful spec than their PC.
| Wavelength | 350-1050nm |
| Spectral Resolution | 0.342nm/ch(2048ch) |
| Integration Time | 0.008-2.04 sec |
| Gain | 1-8 |
| Accumulation Count | 1-255 |
| FOV Angle | 20 deg |
| Quantization Level | 16 bit |
| Sensor Height | 1.5-5m |
| Sensor Azimuth Angle | 0-360 deg |
| Sensor Zenith Angle | 0-70 deg |
Table.1 Measurement System.
| Sensor azimuth angle [deg] | Sensor scan angle [deg] |
| 0,30,60,90,120,150,180,210,240,270,300,330 | 10,20,30,40,45,50,55,60 |
10,20,40,50,70,80,100,110,130,140,160,170,190
200,220,230,250,260,280,290,310,320,340,350 | 50,55,60 |
| 0 | 0 |
Table.2 One Cycle of Measurements
Fig.1 Measurement System
Fig.2 Measurement area
| Time | Date | Count of point | Latitude | Longitude | Vegetation coverage |
| 9:28~19:17 | 98/08/09 | 16 | 46'01.466 | 106'19.935 | 0.15 |
| 11:49~17:23 | 98/08/10 | 11 | 45'36.945 | 105'38.454 | 0.054 |
| 11:16~17:23 | 98/08/11 | 11 | 45'23.300 | 106'14.196 | 0.079 |
Table. 3 Measurement Condition
|
|
|
|
|
|
|
