Á¤º¸°úÇÐȸ ³í¹®Áö B : ¼ÒÇÁÆ®¿þ¾î ¹× ÀÀ¿ë
Current Result Document : 16 / 16
ÇѱÛÁ¦¸ñ(Korean Title) |
ÁöÁöº¤Åͱâ°è¿Í ÀûÀÀÀû Ư¡À» ÀÌ¿ëÇÑ °ÀÎÇÑ Áö¹®ºÐ·ù |
¿µ¹®Á¦¸ñ(English Title) |
A Robust Fingerprint Classification using SVMs with Adaptive Features |
ÀúÀÚ(Author) |
¹ÎÁرâ
Á¶¼º¹è
Jun-Ki Min
Sung-Bae Cho
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 35 NO. 01 PP. 0041 ~ 0049 (2008. 01) |
Çѱ۳»¿ë (Korean Abstract) |
Áö¹®ºÐ·ù´Â Áö¹®À» Àü¿ªÆ¯Â¡¿¡ µû¶ó ¹Ì¸® Á¤ÀÇµÈ Å¬·¡½º·Î ºÐ·ùÇÏ¿© ´ë±Ô¸ð Áö¹®½Äº°½Ã½ºÅÛÀÇ ¸ÅĪ½Ã°£À» °¨¼Ò½ÃÅ°´Âµ¥ À¯¿ëÇÏ´Ù. ±×·±µ¥, Áö¹®ÀÇ °íÀ¯¼ºÀ¸·Î ÀÎÇØ Àü¿ªÆ¯Â¡ÀÌ ´Ù¾çÇÏ°Ô ºÐÆ÷ÇÔ¿¡µµ ºÒ±¸ÇÏ°í, ±âÁ¸ÀÇ Áö¹®ºÐ·ù ¹æ¹ýµéÀº ¸ðµç Áö¹®¿¡ ´ëÇØ °íÁ¤µÈ ¿µ¿ªÀ¸·ÎºÎÅÍ ºñÀûÀÀÀûÀ¸·Î Àü¿ªÆ¯Â¡À» ÃßÃâÇÏ¿´´Ù. º» ³í¹®¿¡¼´Â ´Ù¾çÇÑ Áö¹®À» È¿°úÀûÀ¸·Î ºÐ·ùÇϱâ À§ÇØ °¢ Áö¹®¿¡ ÀûÀÀÀûÀ¸·Î Ư¡À» ÃßÃâÇÏ´Â ¹æ¹ýÀ» Á¦¾ÈÇÑ´Ù. ÀÌ´Â °¢ Áö¹®ÀÇ À¶¼± ¹æÇâÀÇ º¯È·®À» °è»êÇÏ¿© ÀûÀÀÀûÀ¸·Î Ư¡¿µ¿ªÀ» Ž»öÇÑ ÈÄ, Ư¡¿µ¿ª³»ÀÇ À¶¼± ¹æÇâ °ªÀ» Ư¡º¤ÅÍ·Î ÃßÃâÇÏ°í ÁöÁöº¤Åͱâ°è(Support Vector Machines)¸¦ ÀÌ¿ëÇØ ºÐ·ùÇÑ´Ù. º» ³í¹®¿¡¼´Â NIST4 µ¥ÀÌŸº£À̽º¸¦ ÀÌ¿ëÇÏ¿© ½ÇÇèÀ» ¼öÇàÇÏ¿´´Ù. ±× °á°ú 5Ŭ·¡½º ºÐ·ù¿¡ ´ëÇØ 90.3%, 4Ŭ·¡½º ºÐ·ù¿¡ ´ëÇØ 93.7%ÀÇ ºÐ·ù¼º´ÉÀ» ¾ò¾úÀ¸¸ç, ºñÀûÀÀÀûÀ¸·Î ÃßÃâÇÑ Æ¯Â¡º¤ÅÍ¿ÍÀÇ ºñ±³½ÇÇèÀ» ÅëÇØ Á¦¾ÈÇÏ´Â ÀûÀÀÀû Ư¡ÃßÃâ¹æ¹ýÀÇ À¯¿ë¼ºÀ» ÀÔÁõÇÏ¿´´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
Fingerprint classification is useful to reduce the matching time of a huge fingerprint identification system by categorizing fingerprints into predefined classes according to their global features. Although global features are distributed diversly because of the uniqueness of a fingerprint, previous fingerprint classification methods extract global features non-adaptively from the fixed region for every fingerprint. We propose an novel method that extracts features adaptively for each fingerprint in order to classify various fingerprints effectively. It extracts ridge directional values as feature vectors from the region after searching the feature region by calculating variations of ridge directions, and classifies them using support vector machines. Experimental results with NIST4 database show that we have achieved a classification accuracy of 90.3% for the five-class problem and 93.7% for the four-class problem, and proved the validity of the proposed adaptive method by comparison with non-adaptively extracted feature vectors.
|
Å°¿öµå(Keyword) |
Áö¹®ºÐ·ù
ÀûÀÀÀû Ư¡ÃßÃâ
À¶¼± ¹æÇâ º¯È·®
ÁöÁöº¤Åͱâ°è
Fingerprint classification
Adaptive feature extraction
Variations of ridge directions
Support vector machines
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|