ISPRS Commission III, Vol.34, Part 3A ,,Photogrammetric Computer Vision", Graz, 2002 
  
  
  
  
Score 
e 
P 
Lt 
+ 
= c 
opos) 
e 
Weis 
(à 
  
  
0 
02 03 04 05 06 07 08 09 1 val i i : i i 
Visibility 04 02 0 02 04 056 08 
(a) Score vs. visibility 
  
(b) Position errors 
  
  
  
  
  
  
  
0.6 T T T T T 0.6 
04 + i j 0.4 + 
02 } > UNE 1 02r ; 
4 d E 
0 * ti 0 3 + 
> = € 
-0.2 + + + + nf. -02 + > 
-04 L T ri 1 -04 } 
-0.6 i L : L i -0.6 L 4 L i 
-0.4 -0.2 0 0.2 0.4 0.6 0.8 -0.4 -0.2 0 0.2 0.4 0.6 0.8 
(c) Pos. err. adjustment (1) (d) Pos. err. adjustment (3) 
  
  
= 200 | 
di 
To 150 
es | E | 
100 | 
1 
  
Score 
i 
*» 
++ 
| ^ 
i 
i 
  
  
0 
02 03 04.05 06 07.08 09 1 -1 
Visibility 
(e) Score vs. visibility 
  
| 
00 Lia a 
-100-80 -60 -40 -20 0 20 40 60 80 100 120 
(f) Position errors 
Figure 3: Extracted scores plotted against the visibility of the ob- 
ject (a) and position errors of the extracted matches for the pro- 
posed approach using extrapolation (b) and least-squares adjust- 
ment with one (c) and three (d) iterations; scores vs. visibility (e) 
and position errors (f) using the Hausdorff distance. 
A more thorough evaluation of the proposed method, including 
a comparison to a larger number of algorithms can be found in 
(Ulrich and Steger, 2001). 
7 CONCLUSIONS 
A novel object recognition approach for industrial inspection us- 
ing a new class of similarity measures that are inherently robust 
against occlusion, clutter, nonlinear illumination changes, and 
global as well as local contrast reversals, has been proposed. The 
system is able to recognize objects under similarity transforma- 
tions in video frame rate. A performance evaluation shows that 
extremely high object recognition rates (more than 98% in the 
test data set) are achievable. The evaluation also shows that ac- 
curacies of 1/22 pixel and 1/100 degree can be achieved on real 
images. 
REFERENCES 
Ballard, D. H., 1981. Generalizing the Hough transform to detect 
arbitrary shapes. Pattern Recognition 13(2), pp. 111-122. 
Borgefors, G., 1988. Hierarchical chamfer matching: A para- 
metric edge matching algorithm. IEEE Transactions on Pattern 
Analysis and Machine Intelligence 10(6), pp. 849—865. 
A - 350 
  
"Pos. err. extrapolation T 
Pos. err. adjustment (1) 
Pos. err. ádjustment, (3) 
     
  
    
  
  
  
0 5 10:-.15 20 25 30/35. 40. 45 50 
(a) Position errors 
  
0.08 
  
004} | [4] 
  
  
i A 
if Vl 
-0.06 + | Angle err. extrapolation 4 
Angle err. adjustment (1) -------- 
0.08 , Angle etr. adjustment (3) ee 
  
0 § 10 315. 20 25-30 35 40 45 50. 55 
(b) Orientation errors 
Figure 4: Position errors of the extracted model poses (a); orien- 
tation errors of the extracted model poses (b). 
Brown, L. G., 1992. A survey of image registration techniques. 
ACM Computing Surveys 24(4), pp. 325-376. 
Canny, J., 1986. A computational approach to edge detection. 
IEEE Transactions on Pattern Analysis and Machine Intelligence 
8(6), pp. 679—698. 
Fórstner, W., 1994. A framework for low level feature extraction. 
In: J.-O. Eklundh (ed.), Third European Conference on Computer 
Vision, Lecture Notes in Computer Science, Vol. 801, Springer- 
Verlag, Berlin, pp. 383—394. 
Lai, S.-H. and Fang, M., 1999. Robust and efficient image align- 
ment with spatially varying illumination models. In: Computer 
Vision and Pattern Recognition, Vol. II, pp. 167-172. 
Olson, C. F. and Huttenlocher, D. P., 1997. Automatic target 
recognition by matching oriented edge pixels. IEEE Transactions 
on Image Processing 6(1), pp. 103-113. 
Rucklidge, W. J., 1997. Efficiently locating objects using the 
Hausdorff distance. International Journal of Computer Vision 
24(3), pp. 251-270. 
Steger, C., 1998. An unbiased detector of curvilinear structures. 
IEEE Transactions on Pattern Analysis and Machine Intelligence 
2002), pp. 113-125. 
Steger, C., 2000. Subpixel-precise extraction of lines and edges. 
In: International Archives of Photogrammetry and Remote Sens- 
ing, Vol. XXXIII, part B3, pp. 141—156. 
Tanimoto, S. L., 1981. Template matching in pyramids. Com- 
puter Graphics and Image Processing 16, pp. 356—369. 
Ulrich, M. and Steger, C., 2001. Empirical performance evalua- 
tion of object recognition methods. In: H. I. Christensen and P. J. 
Phillips (eds), Empirical Evaluation Methods in Computer Vi- 
sion, IEEE Computer Society Press, Los Alamitos, CA, pp. 62- 
76.