GISdevelopment.net ---> AARS ---> ACRS 1989 ---> Poster Session

Computer - Assisted HILL - Shadding Generation from DEM for cartographic enhancement on thematic maps

Raiadika Mastra
National mapping of Indonesia

Hiroaki Kakihchi,Shunji Murai
Institute of Industrial Science
University of Tokyo, japan.


Abstract
There is no doubt that a good eartographic presentation on a map enhances the readability of the map itself. The benefits from the user of the computer as a tool for such enhancement is considerable. Especially in countries where the labor cost is high. Up till recently, the technique widely used to produce hill - shading presentation on a map has been hand drawing or air-brushing.

In this paper, the use of the computer to generate the hill-shading image from DEM data bank is introduced. Results show that this maps and other small scale maps. At the same time it. Can reduce drawing time and other cartographic works.

Introduction
Automation in cartography is not only involves a technical technological and scientific aspect. It also presents human aspects, which are essential to ensure its success and its transition from an experimental stage to an operational one. Also, automated cartography emphasizes the role to eartographers as designers.

Many parts of the cartographer's works can be automated. Hut discussions in this paper emphasizes more the creating of shading effect for the small scale maps or other thematic maps.

The 3-D effect on a map is very helpful to readers who have difficulties in sceing the general view of terrain. As for ordinary users, atlases, thematic maps and even to their Topographic maps. The most famous country to apply the hill-shading effect to their maps is the swiss. Most of cartographers will agree that the Swiss maps are one of the finest cartographic works in the world. Many countries which are applying this technique are still depending on their skilled cartographers who are trained to use air brushing or hand-drawing. To produces the hill - shading effect on one map sheet needs a lot of time. Patience and highly skilled cartographers. However, these can be replaced by using he computer as a tool to generate hill - shading image.

Result shows it is applicable to use this shading image to apply on maps and furthermore overlayed on other features to make a complete themachic map. Tesult shows it is applying layer-slicing with special assigned colours on each layers, the topography of the map itself will appear nicer and also will give a stronger effect.

The use of Hue ( r, g, b), Saturation and Intensity (HIS) instead ofR.G>B. colours only, for producing choropleth image and the combination of shade image with HIS to produce colored image for final use and CRT display is introduced. To give a general view, a brief example of costing aspects for producing this image is also discussed.

Methodolgoy
The G.S.I. ( Geo-graphical Survey Institute) of Japan has already stored a Digital Elevation Data bank of the whole country and in the form of 250 meter mesh data. This data was used to create the DTM and shaded DTM according to desired direction of sunrays to give a 3-D effect.By the availability of this data in CCT format, it is quite possible to use it as feeding data to create a hill - shading image and chorophlth image.

The following diagram shows the general steps of work and how the hill- shading and other output products have been made.


Shaded Image.
  1. Shading Image
    The best 3-D effect of topography on a map where the human eyes can perceive is by placing a light source at the north-west corner of the map sheet and made an angle of 45 degree to the upper edge of the map sheet. The result will gives a nice 3-D perception of the topography itself.

    There are many algorithms available for creating the hill - shading image, however, in this case a simple equation was applied to the DEM data.

    The following equation shows the relationship between shade and illumination sources .

    S = 1000 * Cos A * Sin B

    Where:
    S: Shaded image
    O<S<100
    S = 1000 bright
    S = O dark

    A: Angle subtended by the direction fo sun ray and the normal
    B: Angle substended by the normal and the vertical



    Figure 1. Configuration of angle.

    By applying the above algorithm to the DEM data. This will give a grayish hill - shading image with the gray values ranging between o to 100. This value's range can be changed accordingly ( Figure 2a).

  2. Hill - Shading image from Slope Aspect
    If the appearance of the shade effect from the previous method need to be more stronger, another simple formula can be use. This formula is actually use for creating slope aspect image. And by assuming that the light source comes from the same direction as the previous methods, sharp edge and contrast image can be produced (Figure 2B).

    If the following formula shows the relationship between slope aspect and elevation mesh data.

    If 0 is the slope aspect angle measured clockwise from the North. Some conditions can be introduced as follows:


    Figure 2. Elevation mesh data.
Choropleth image
To make the map appear nicer as well as easier to distinguish, a choropleth image is made. The choropleth image is simply made by slevation data which has valueranging from o to 100 meter is assigned to one group and one colour. The same is applied to another group of data. Finally a colourful choropleth image is made ( figure 3. )

Combined images.
After both hill - shading image and choropleth image have been made, the next step is to combine these two images into one file. This combined file image is used to test the intensity of the shading file. To check if the gray level of shade area is light enough and does not block the other images . By replacing the I Instensity) value of HIS with hill-shading image, the combined image is then produced ( figure 4).

At this stage other information related to the final product can be combined directly. With all these information available, the final map can then be plotted using laser printer on laser graphic film and can later be printed on ordinary offset printing press.

Cost aspect
A direct calculation varies from one company to another but as an illustration from one of the mapping companies, the rough calculation for producing one sheet 100 cm * 75 cm map with 3 colours from the existing DEM data, hill-shading generation and to plot the result on laser graphic film by laser printer will cost about Y 40, 000 to Y 50, 000 per sheet. These films are the final sheets and can be used directly for printing purposes.


Figure 2: The hill-shading image method(shade aspect)


Figure 3: Choropleth image over part of
Honshu island of Japan
Figure 4 Combined image

Figure 5a: The image of Hokkaido is
generated form computer
Figure 5b:From conventional hill-shading technique

Conclusion
The test study showed that computer generated hill-shading can meting the requirements for addition of height information and as an art product. Production of this image need only a few minutes of CPU time.

For atlases and other small scale thematic maps, the first method is more suitable. This is due to the very soft appearance of the relief image, but for larger scale maps with mountainous topographic surfaces, the slope aspect technique is more suitable.

For flat terrain area with or relatively equal proportion of flat and hilly area, the first method is more applicable. The disadvantage of the second methods, is that the slope aspect formula does not fit to areas with small relief differences.

If the majority of the mountain configurations form the North-East slope and South-West slope, the 3-D effect of topography on a map will give no sharp appearance. This is due to the same amount of light received by the mountain redges. To overcome this disadvantages, the light source need to be shifted a little West instead of just placing a light source at the north-west corner of the map sheet. This will give the same 3-D effect.

Due to working directly on the computer, the time-consuming work usually encountered by cartographers to make hand-drawings or air-brushing no longer exists. Digital drawing is certainly more flexible and efficient compared to manual work. The costing illustration which is mentioned above cannot be judged as a fixed price since only rough estimation is available.

References.
  • E. Kmhof, Kartographische Gelande-Darstellung, Walter de Gruyter & Co. 1965.
  • Grelot, Jean-Phillippe, Human aspects in Cartography Automation, 1986 ACSM-ASPRS Annual Convention Vol..1, Cartography and Education.
  • J.I.C.A. (Japan International Cooperation Agency) . The Remote Sensing Engineering Project for the Development of Agricultural Infrastructure in the Republic of Indonesia, March, 1983.
  • Tateishi Ryutaro, DTM manual ( in Japanese(, Mural Laboratory.