INDEX, MIDDLE, AND RING FINGER EXTRACTION AND
IDENTIFICATION BY INDEX, MIDDLE, AND
RING FINGER OUTLINES
Ching-Liang Su
Department of Industrial Engineering and Technology Management, Da Yeh University
112 Shan-Jiau Road, Da-Tsuen, Chang-Hua, Taiwan 51505, China
Keywords: Finger Shape Identification, Finger Shape Extraction, Locating Fingertip and Finger-to-Finger-Valley,
Geometry Matching.
Abstract: In this study, the new technique is used to extract the index, middle and ring finger outlines. The
orientations and geometrical features of these outlines are calculated and compared to identify different
individuals. The techniques of database SQL searching and manipulation, image dilation, object position
locating, image shifting, rotation, and interpolation are used to recognize different individuals. The hand
was fixed each time when a photograph was taken, and one can assume that each time when a hand was
acquired, the image was the same as the previous one. Since the photographs are the same, after the index,
middle or ring fingers have been extracted from the hand image, the acquired images can be used to identify
different persons.
1 INTRODUCTION
In this study, the new technique is used to extract the
index, middle, and ring finger-outlines and to
perform the person’s identification. The image
automatic registration technique locates the
orientations and positions of the extracted finger
images. In order to perform the geometrical
comparison of two finger images, the centroids of
the finger images need to be located. The images
used in this study are the 128 by 128 images. One
needs to perform the image movement to move the
image to the center of the frame. Furthermore, the
major axes of the finger images need to be located
and we need to move the images to allow the major
axes of the finger images to be in the straight
position. Since every image is centered and
straightened, one can perform the image geometrical
comparison to identify different fingers. The above
steps are performed by computer itself and no further
human involvement are required. The algorithm
developed in this study can identify fingers.
2 ACQUIRING HAND IMAGE
Figure 1 shows the acquired hand images. After
further processing, one can obtain the hand-edge
images. By the edge thinning direction and the hand
geometry features, one can extract the fingertips and
finger-to-finger-valleys, as shown in figure 2.
3 EXTRACTING INDEX,
MIDDLE, AND RING FINGERS
After the algorithm locate the locations of the
fingertips and finger-to-finger valleys, one can
extract the index, middle, and ring finger. The
extracted results are shown in figures 3, 4, and 5,
respectively. After further processing one can obtain
the fully closed finger images.
Figure 1: Acquired hand images.
518
Su C. (2008).
INDEX, MIDDLE, AND RING FINGER EXTRACTION AND IDENTIFICATION BY INDEX, MIDDLE, AND RING FINGER OUTLINES.
In Proceedings of the Third International Conference on Computer Vision Theory and Applications, pages 518-520
DOI: 10.5220/0001079505180520
Copyright
c
SciTePress
Figure 2: Extracted fingertips and finger-to-finger valleys.
4 IMAGE REGISTRATION AND
SUBTRACTION
In order to perform the comparison of two fingers,
the centroids of the finger images need to be
determined. The images used in this study are the
128 by 128 frames. One needs to perform the image
movement to move image to the center of the 128 by
128 frame; i.e. one needs to shift image to allow the
centroid of each image to be in location (64,64) of
the frame. Furthermore, major axes of fingers need
to be located and we need to move fingers to allow
major axes of the fingers to be aligned to the same
straight position. Since every image is shifted to the
center of the 128 by 128 frame and the major axis of
each image is aligned to the same vertical position,
one can perform the image subtraction to determine
the differences of each two fingers.
After the image movement, as mentioned
previously, in the new feature-domain, every finger
image would be aligned to the same straight and
center position.
After image rotating, shifting, and interpolating,
two finger images overlap. Since both fingers
overlap, image subtraction can now be applied to
compare the difference of these two fingers. By the
subtracted results, one can identify different fingers.
Figure 6 shows the image subtractions of index
fingers. The subtracted-data of the index fingers is
shown in table 1.
5 RESULTS AND CONCLUSIONS
Table 1 shows the partial comparison data on
differences after performing image subtraction. The
data inside the rectangular boxes in the table are the
results of subtraction for genuine index finger. The
other data in this table comprise the subtracted
results from two different index fingers. By
analyzing the date in table 1, one can conclude that
the accuracy rate is 97%.
ACKNOWLEDGEMENTS
The National Science Council, Taiwan, supported
this work under grant NSC 96-2221-E-212-004.
REFERENCES
Miguel A. Ferrer, Carlos M. Travieso, and Jesus B.
Alouso, “Using hand knuckle texture for biometric
identifications,” IEEE A&E Systems Magazine, June
2006
Sotiris Malassiotis, Niki Aifanti, and Michael G. Strintzis,
“Personal Authentication Using 3-D Finger
Geometry,” IEEE Transactions on Information
Forensics and Security, vol. 1, no. 1, March 2006
Erdem Yörük, Ender Konuko˘Glu, and Bülent Sankur,
“Shape-based hand recognition,” IEEE Transactions
on Image Processing, vol. 15, no. 7, July 2006
Figure 3: Extracted index fingers.
Figure 4: Extracted middle fingers.
INDEX, MIDDLE, AND RING FINGER EXTRACTION AND IDENTIFICATION BY INDEX, MIDDLE, AND RING
FINGER OUTLINES
519
Figure 5: Extracted ring fingers.
Figure 6: Index finger subtraction.
Table 1: Comparison-data of index fingers.
Comparisons of same
person’s index fingers
13606 9344 32648 35920 32057 19731 19951 20553 21518 22666 24366 23012 23208 24767
9075 36050 39984 35514 16802 18862 17742 27583 27046 25050 19072 19944 23275
32330 36490 31081 17812 20023 17886 21973 22145 22144 21070 21091 23797
15355 6971 39351 40742 41120 27115 28159 34670 42610 45765 41568
14295 45178 45510 47317 29593 29584 34317 45483 49224 43513
39852 41553 40999 27351 26778 32646 42803 45542 42197
8055 11227 28340 28836 28194 12836 12500 17419
11585 29746 31098 30431 14746 15300 18198
30076 31020 30898 16496 12981 20415
11496 18355 32092 34529 31320
14401 32070 34369 31665
28564 31604 27848
10206 11121
14594
VISAPP 2008 - International Conference on Computer Vision Theory and Applications
520