/[gzz]/manuscripts/Paper/exper.txt
ViewVC logotype

Diff of /manuscripts/Paper/exper.txt

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1.1 by jvk, Tue Feb 25 14:36:02 2003 UTC revision 1.2 by jvk, Tue Feb 25 17:15:58 2003 UTC
# Line 7  Test cases: Line 7  Test cases:
7       - white       - white
8       - one color       - one color
9       - full texture       - full texture
10       - grayscale texture       - grayscale texture (intensity scale "compressed" to a certain range)
11    
12  General:  General:
13    
# Line 15  General: Line 15  General:
15  - the order of the test cases is random for each subject  - the order of the test cases is random for each subject
16    to componsate memory effects    to componsate memory effects
17    
18  Recognisability test (usefulness):  Recognisability test (usefulness)
19    ---------------------------------
20    
21  - objective: show that the background textures are useful  - objective: show that the background textures are useful
22  - means: show that fragments of documents can be recognized much  - means: show that fragments of documents can be recognized much
# Line 25  Recognisability test (usefulness): Line 26  Recognisability test (usefulness):
26    
27    
28    
29    
30  Suppose that the first page of each of hundreds of documents is torn into  Suppose that the first page of each of hundreds of documents is torn into
31  pieces and the pieces are put in a single container.  pieces and the pieces are put in a single container.
32  It is intuitively clear that the pieces can be identified and put  It is intuitively clear that the pieces can be identified and put
33  back together much more rapidly if each document has a unique  back together much more rapidly if each document has a unique
34  background texture as compared to, e.g., single background colors.  background texture as compared to, e.g., single background colors.
35    
36  Another question is whether the extra features of textures help  Another question is whether the extra features of the textures help
37  recognition as much if the fragments cannot be seen simultaneously  recognition as much if the fragments cannot be seen simultaneously
38  for comparing.  for comparing.
39    
40    "Applicability"(better word?) test
41    ----------------------------------
 "Applicability"(better word?) test:  
42    
43  - objective: show that the textures are better than single colors  - objective: show that the textures are better than single colors
44    
# Line 49  for comparing. Line 50  for comparing.
50  XXX: small-scale fragments do lose shape features.  XXX: small-scale fragments do lose shape features.
51  what to say about this  what to say about this
52    
53  Readability test (non-disruptiveness):  Readability test (non-disruptiveness)
54    -------------------------------------
55    
56  - objective: show that readablity does not suffer much  - objective: show that readablity does not suffer much
57  - means: show that text size has much stronger effect on readability  - means: show that text size has much stronger effect on readability
58    than the presence of background texture    than the presence of background texture
59    
60       - 4 test cases: small/big font with white/texture bg      - 2 test cases: with white bg or texture bg
61        - each subject reads a number of (perhaps 4) texts on the two types of
62       - each subject reads a number of (perhaps 4) texts on different        backgrounds and the time is measured
63         backgrounds and sizes and the time is measured      - font size is varied between texts and/or subjects
64          (or should the same text be read in different sizes?)
65       - the order of the test cases is random for each subject to      - is the memory effect stronger than the variation between different texts?
66         componsate memory effects      
   
           - 3 subjects should do small textures  
             first and the other 3 subjects big textures first  
             (not sure about this; the reasoning is that the  
             effect of the variation between subjects on white/texture  
             discrimination is lessened at the expense of  
             small/big discrimination which should be clear enough anyway)  
   
   

Legend:
Removed from v.1.1  
changed lines
  Added in v.1.2

savannah-hackers-public@gnu.org
ViewVC Help
Powered by ViewVC 1.1.26