approach instead of a top to bottom which is the 
conventional way of teaching process. The course 
material, however, is organized as usual into indi- 
vidual modules or sections and each section dealing 
with a specific problem, then the BTB approach is 
applied to that section as follows: The instructor 
introduces in each section or chapter a group of 
routines which deal with representative problems 
within the chapter. Those problems become individ- 
ual lab exercises which require the development of 
an algorithm and the writing of a computer program. 
The algorithm and the running program have already 
been prepared by the instructor. First the 
instructor demonstrates the way the program works 
as an operator (Fig. 1) and allows the students to 
work with it so that to create strong motives to 
them to study the scientific bases. At the same 
time it is given a lab exercise which requires the 
development of an algorithm and the creation of a 
running program. Parallel to the lab assignment the 
scientific bases are analysed in the lecture and 
the corresponding algorithms are developed. During 
the lecture there are many questions from the stu- 
dents because they are working on the lab assign- 
ment and they need to completely understand all the 
scientific bases to finish the program. 
The instructor, for each lab exercise, may provide 
the source code of several routines or may provide, 
with certain restrictions, the source code listing 
of the entire program, so that the average student 
will finish the assignment within the expected time 
limits. The lecture does not have to enter into a 
great depth of the theory leaving that option to 
the interest of those who want to study further 
from provided references. In this way a great 
amount of course material is covered and there is 
an integrated learning process of all components 
involved: theory (at any depth), algorithm, soft- 
ware, hardware plus that the student is confident 
of knowing the content of the study section. This 
learning process was applied first during my nine 
year career at the Surveying Engineering Program of 
the California State University, Fresno and pro- 
vided excellent results (Hatzopoulos J.N. 1985). 
In the Department of Environmental Studies, of the 
University of the Aegean, the same process is also 
applied for remote sensing and GIS courses. There 
is a microcomputer laboratory with PC compatible 
microcomputers having CGA and Hercules graphics 
cards. This Tab is used for both remote sensing lab 
assignments and GIS lab assignments. Two examples 
one in remote sensing and one in GIS of lab assign- 
ments are given bellow: 
Remote Sensing Lab # 11. RADIOMETRIC ENHANCEMENT 
A part of Landsat image of channel 4 composed of 
512 lines by 512 columns is available in the file 
M. DAT. Write a computer program to do the follow- 
ing: 
l. Generate 13 gray scale Shades by making a hyper- 
pixel composed of 3x4 = 12 single pixels of the 
PC-CGA graphics card which will be worked in mono- 
chrome high resolution graphics mode. 
2. Read the available file M4.DAT one line at a 
time and hold in memory those values needed to 
create the regular histogram of the digital image 
and the equalized histogram. 
3. Create a menu within the program which allows 
the following operations for radiometric enhance- 
ment and image display: 
(a) Display regular histogram. 
(b) Display equalized histogram. 
(c) Display stretched histogram. 
134 
(d) Display regular image. 
(e) Display equalized image. 
(f) Display stretched image. 
(g) Change image stretch values. 
(h) Define upper left corner coordinates of image 
display window 60x150. 
Notices 
The following files are provided to you in the 
microcomputer lab: 
IMPRO1.BAS: contains the source code of a program 
which solves the given problem. 
IMPRO1.EXE: binary version of IMPRO1.BAS. 
M4 .DAT: The channel 4 of Landsat-TM data from 
the area of Mytilene. 
Wanted (a) The program list with source code which 
will have different variable names from 
those in the IMPRO1.BAS and from other 
students. 
(b) A copy of the regular, equalized and 
stretched histogram in a dot matrix 
printer using PrtSc command. 
(c) A copy of the regular, equalized and 
stretched image in a dot matrix printer 
using PrtSc command. 
(d) A written report on the impact of the 
radiometric enhancement (equalization, 
stretch) to identify: sea water, build- 
‘ings, the soccer fields, the airport, 
rangeland, agriculture and forest land. 
The graphic outputs of this assignment are shown in 
Figures 2, 3, 4, 5, 6, and 7. 
  
Figure 2. Regular histogram of M4.DAT image. 
An 
jaw 
Aw 
    
   
   
SAC 
MAMMA Y RAR ARTA AR AAAAA AAA 
he Hi it (RETURN): MAMAMMMAMMMMMAMAMAMAM fii 
i jn MMAMMAM 3333424448 MA SAAAANAAMAAAANARAARAAN) JARAARS 
n HAMM IH 237 , HAMA MAMMA ARAAAAN 
fhe A AA i fifi U an Maa a 
MOM i 
jm i UE 4 n nama 
HM MN E ann E 
im AAMMARUAT 3 IARAANANAA, dM MAMMAARARARAIGL ) AMITAAR SARI 1 RAR 
A e MAMMA A 17 IARAAARARANANARARA VIL] ARAM y VIAM 
nn MM? EE A n ARAMRARARARAMU y JANRILLLANRALL) 
2 pretty aa i , MAMAMMMAM UN] 12) 
A 
" i E EUM EX Be A MIRA 
M E una, m AMAN d ds 
D 
M 
a 
4 
AMA M MI AM 
AMAMMM ARRA Y 
               
   
       
      
   
   
  
   
  
   
  
   
      
MH 
Ren bi ar "m zin an 
y MM i "m Sin ME 
ant 
MI 
   
       
  
  
POLE 
NO an 
ati i nri Mes) b p DAMM 
US amis a) nds TAAAMAAMAAARANAMAA (t 
PIMMMAMMII 1A AAARARRARAR) YY 717’ 
fit pra 
^ 
jn MONS TAS jy MNA 
Figure 3. Display of M4.DAT regular image.