Generated Reference Ground Control Points from SAR Image
SAR's GCPs Extraction
The synthetic aperture radar (SAR) is an active sensor and con be used ot observe the earth in the day and night by regardless the weather . the microwave wavelength propagates through the cloud, it has the higher resolution than the ordinary radar according to the synthetic aperture and pulse compressing technology. Radar transmits the pulse of microwave power to the target and records the reflected signal. This action occurs in geometrical form for recording the side looking vies. The energy of back-scattering signal extremely depends upon the incident angle and eh direction of receiving antenna. All information can be used to build up the SAR image. The back-scattering has two position as follows ;
-
object surface area or surface scattering, and
- object inside-volume or volume scattering
The surface scattering means the action continuously occurs at the surface or contacting with te atmosphere such as waster resources ( river, sea, lake ) or solid material that the microwave cannot penetrate through these materials . it is obviously shown on SAR image, especially with black-smooth -surface object ( the reflected energy is very small ), such as river, canal , asphalt road and etc. it should be noted that the surface status of object will assign the index of refraction.
The absorbed energy type of object depends on the dielectric constant and it is an important factor reflected wave such as dry soil. It has an little effect with the microwave propagation. On the other side for living area such as bridge, building, concrete road which are constructed from steel and concrete, the dielectric constant is small. The reflection is high and the white color appears clearly on SAR image. Therefore, the black color appears on SAR image for river, canal, asphalt road but the white color appears on SAR image for building, concrete road. Then, the generate reference ground control points(GCPs) cane be easily done .
Fig .3 shows an SAR image of upper area of Bangkok metroplitan
It is obviously sown that the Chao Pharya river , canal and asphalt road are in black color . the concrete road is in white color. Therefore, to assign the position of GCPs cab e obtain form the road intersection, water intersection, it is necessary to invert the image color for river, canal and asphalt road to be white color by using linear gray scale inversion as in equation (1).
X
min , X
max = lower and Upper range of input image
Y
min - Y
max = Lower and Upper range of output image
Figure 5 shows the image after gray scale inversion, both images must be exclusively together in together tot get rid of the background and the unwanted information. In the image, there are different color appearing on the image , for the other information, the adjacent color are appearing on the image, for the other information, the adjacent color are suppressed as shown in figure 6. it is obviously that the river and canal will be white color, thus the intersection of road and river ( or canal ) will be assigned as typical GCP for transforming to be the other image.
Fig. 5 the result image obtained by using gray scale inversion method
Fig. 6 Image after exclusion between original and gray scale inversion image.
Fig. 7 Received OPS from JERS-1 on 16 January 1998 for the same area of SAR (Fig.3)
