The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part BI. Beijing 2008 
As introduced above, in order to conduct the fusion of points 
cloud and images, the points should be interpolate into raster 
format firstly. Interpolating can be regarded as resample. Some 
approaches, such as Inverse Distance Weighted(IDW), Spline, 
Kriging can be used here. In this paper, we use Inverse 
Distance Weighted method to interpolate the vector points into 
raster. 
As mentioned in section 2.2, Z coordinate and the intensity are 
included in points data. Both of the two data reflects the 
objects information. So, during the interpolation, Z and the 
Intensity values are used as the key value. Fig.3 and Fig.4 
indicates the results of interpolation by Z and intensity values. 
Fig.3 Raster Image by Interpolation of Z 
Fig.4 Raster Image by Interpolation of Intensity 
Since some uncertainly factors and different spectral 
characteristic during the data acquisition, the intensity values is 
not as normal as Z values. Even in the same building, the 
intensity will change violently. Thus, Fig.3 is more smoothness 
and flatness than Fig.4. The buildings and the hills which are 
higher than the terrain is quite obvious in Fig.3 but the 
buildings can not be recognized in Fig.4. However, the road 
and the plants is clear and vivid compared with the Fig.3. 
3. DATA FUSION 
3.1 Raster Fusion between image and points cloud 
IHS fusion 
The IHS colour space is broken down into Hue, Saturation and 
Intensity. In order to separate the intensity from IHS colour 
space from the intensity value of points, we use / to denote 
the intensity value of IHS colour space and Intensity to 
denote the intensity value of points cloud. Hue refers to pure 
colour, saturation refers to the degree or colour contrast, and 
intensity refers to colour brightness. Modeled on how human 
beings perceive colour, this colour space is considered more 
intuitive than RGB. It can be compared to the dials on an old 
television set that help viewers adjust the set's colour. 
To analyze and process images in colour, machine vision 
systems typically use data from either the RGB or HSI colour 
spaces, depending on a given task's complexity. For example, 
in simple applications such as verifying that a red LED on a 
mobile phone is indeed red and not green, a vision system can 
use data from R, G and B signals to perform the operation. 
With more complex applications, however, such as sorting 
pharmaceutical tablets of subtly different colours, a vision 
system may require hue, saturation and intensity information to 
perform the operation. 
IHS fusion is based on the conversion between IHS colour 
space and RGB colour space. It is useful on the fusion 
between multi-spectral images and the panchromatic images. 
During the IHS fusion process, the RGB values of multi- 
spectral image should be converted to IHS values for each 
pixel. Since there’s only grey value in panchromatic images, 
the grey value is considered as the RGB values. So, the 
panchromatic images can also be converted to IHS colour 
space. Then the I values of multi-spectral image can be used 
for further process by some other image such as panchromatic 
images. Fusion image can be get after the inverse transform of 
IHS colour space to RGB. 
The first fusion experiment is executed between Ortho 
photomap and results of interpolation of Z value. Here, the / 
values of Ortho-photomap are replaced by the grey value of 
interpolation results. Fig.5 shows the result of the first 
experiment. 
The second experiment is executed for Ortho-photomap and 
results by Intensity value interpolation. Here, the I values 
of Ortho-photomap are replaced by the grey value of 
interpolation results. Fig.6 shows the result of the second 
experiment. 
311