Color 
© 
"t 
un 
= 
g 
inction 
  
| 
ct image 
each fixed 
only one 
djustment 
cause the 
ixed color 
rrounding 
tion. The 
Space is 
ing steps. 
lor space. 
etermined 
spondence 
steps are 
in HSI 
(2) 
‚hen fixed 
0) 
(3) 
on m, {he 
sed in this 
research. 
m(S) = f(|S-S1|,|S-S2|.....,|S-Sn) (4) 
Adjust function which output the adjusted color S’ for 
arbitrary color S, by solving the equation (4) as the 
boundary condition (3). 
Color adjustment by using adjust function. Color 
adjustment was conducted for all pixels in object 
image. 
EXPERIMENT 
Object Images 
Object images used in this study are SPOT PA image 
and Landsat TM image (band 2, band 3, band 4). False 
color image was made by assigning R,G,B to Band 4, 
Band 3, Band 2 respectively. SPOT PA image and 
Landsat TM image were collected on August 1991 and 
November 199] respectively. Object area contains 
kanazawa city and suburbs of kanazawa. This area 
contains urban area, roads, rivers, rice fields, forest, 
sea. Forest area which is very important in this study 
occupied about 20% in object image. The size of object 
image is 1024 x 1024 pixels. 
Evaluation of the fused image. 
Following two methods are selected as the index of the 
evaluation of the fused image. 
(1) Visual interpretation scores. 
(2) Correlation between the Panchromatic channel 
and the derived intensity. 
First method was widely used for the research of fused 
images. Second method is the correlation between the 
density level of original images (Iold) and the density 
level of adjusted image (Inew). 
SUMMARY 
In order to solve the problems which the exist method 
has satellite image fusion, color adjust method was 
introduced in the modification of color space. By using 
this method, color adjustment was conducted for only 
the Decessary area, the area which is not necessary for 
the color adjustment can keep original density level. 
307 
Therefore evident changes of density levels in specific 
area caused by replacing intensity levels obtained by 
exit method could be corrected. Proposed method has 
the advantages that the adjustment is not unnatural for 
the fused image by modifying the color space. In this 
study, one pair of images are tried by using proposed 
method. Same result can be expected for other images. 
Color adjustment steps should be improved in future 
research. 
REFERENCES 
W.Joseph Carper, Thomas M. Lillesand, and Ralph W. 
Kiefer, 1990a. The Use of Intensity-Hue-Saturation 
Transfermations for Merging SPOT panchromatic and 
Multispectral Image Data. Photogrametric Engineering 
& Remote Sensing, 56(4) pp. 459-467 
Jeff R. Harris and Richard Murray, 1990b. IHS 
Transform for the Integration of Radar Imagery with 
other Remotely Sensed Data. Photogrammetric 
Engineering & Remote Sensing, 56(12), pp.1631- 
1641. 
Manfred Ehlers, 1991a. Multisensor Image Fusion 
Techniques in Remote Sensing, ISPRS Journal of 
Photogrammerty and Remote Sensing, 46 pp. 19-31 
Pat S. Chavz, Jr, Stuart C. Sides, and Jeffery A. 
Anderson, 1991b. Comparison of Three Different 
Methods to Merge Multiresolution and Multispectral 
Data: Landsat TM and SPOT  Panchromatic, 
Photogrammetric Engineering & Remote Sensing, 
57(3), pp. 295-303 
A.H. J. M. Perremans, R.W.L. Jordans, and R. 
Allewijn, 
1993a. Merging Multispectral and Panchromatic 
SPOT Images With Respect To Radiometric Properties 
of The Sensor. 59(1), pp. 81-87 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B7. Vienna 1996