In: Wagner W„ Székely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Voi. XXXVIII, Part 7B 
533 
By analyzing the result showed in Figure 4 it can be noticed that 
the highway present continuity loss. This happened due to the 
presence of trees in the borders tracks in the scene acquisition 
instant, hiding information regarding the plan of the same ones. 
This way, it was necessary to accomplish a reconstruction of the 
tracks in order to obtain them with its real plan. Then, the 
morphologic opening operator was applied through the structuring 
element mmseline (line mask). For the use of this structuring 
element was necessary to define two parameters, the line 
thickness and orientation, like this, with base in the accomplished 
tests, it was defined the values 40 and 125 respectively. The 
obtained result is exhibited in Figure 6. 
Figure 6. Reconstruction of the tracks 
For the highway tracks detection, the image was thinning and the 
result submitted to the pruning operation. The pruning consisted 
of removing the segmentation left by the thinning process. The 
result is shown in Figure 7. 
The next step was to apply a dilation on the image through the 
structuring element mmsebox (square mask), seeking to obtain the 
tracks with its original thickness. The Figure 8 presents the 
obtained result. 
Figure 8. Image dilated 
Analysing the Figure 8, it is verified that the employed 
morphologic processing was adapted for the highway detection in 
high resolution image. The quality of the obtained registration 
proves the efficiency of the Mathematical Morphology technique 
in the area of Cartography seeking to the future cartographic 
products updating. With regard to the application of 
morphological operators, the work could be considered as 
concluded, presenting as the best result, as already explained, the 
result obtained in Figure 8. 
3. CONCLUSIONS 
The experiences accomplished with the tools of the Mathematical 
Morphology, with intention of reaffirming its use potential in the 
area of Cartography, they were well happened. It is observed that 
the obtained final product assisted to the objective proposed in to 
demonstrate the viability of the morphologic tools in the process 
of features detection in high resolution images, with view the 
cartographic products updating. 
The key for the obtaining of good results lives in the appropriate 
choices of the structuring elements and thresholds values for the 
several stages of the detection process. For so much, it is