Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ
ÇѱÛÁ¦¸ñ(Korean Title) |
´ÙÁß ºÐÆ÷ ÇнÀ ¸ðµ¨À» À§ÇÑ Haar-like Feature¿Í Decision Tree¸¦ ÀÌ¿ëÇÑ ÇнÀ ¾Ë°í¸®Áò |
¿µ¹®Á¦¸ñ(English Title) |
Learning Algorithm for Multiple Distribution Data using Haar-like Feature and Decision Tree |
ÀúÀÚ(Author) |
Ju-Hyun Kwak
Il-Young Woen
Chang-Hoon Lee
°ûÁÖÇö
¿øÀÏ¿ë
ÀÌâÈÆ
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 02 NO. 01 PP. 0043 ~ 0048 (2013. 01) |
Çѱ۳»¿ë (Korean Abstract) |
Adaboost ¾Ë°í¸®ÁòÀº ¾ó±¼ÀνÄÀ» À§ÇÑ Haar-like featureµéÀ» ÀÌ¿ëÇϱâ À§ÇØ °¡Àå ³Î¸® ¾²ÀÌ°í ÀÖ´Â ¾Ë°í¸®ÁòÀÌ´Ù. ¸Å¿ì ºü¸£¸ç È¿À²ÀûÀÎ ¼º´ÉÀ» º¸ÀÌ°í ÀÖÀ¸¸ç ÇϳªÀÇ ¸ðµ¨À̹ÌÁö°¡ Á¸ÀçÇÏ´Â ´ÜÀϺÐÆ÷ µ¥ÀÌÅÍ¿¡ ´ëÇØ ¸Å¿ì È¿À²ÀûÀÌ´Ù. ±×·¯³ª Á¤¸é ¾ó±¼°ú Ãø¸é ¾ó±¼À» È¥ÇÕÇÑ ÀÎ½Äµî µÑ ÀÌ»óÀÇ ¸ðµ¨À̹ÌÁö¸¦ °¡Áø ´ÙÁß ºÐÆ÷¸ðµ¨¿¡ ´ëÇؼ´Â ±× ¼º´ÉÀÌ ÀúÇϵȴÙ. ÀÌ´Â ´ÜÀÏ ÇнÀ ¾Ë°í¸®ÁòÀÇ ¼±Çü°áÇÕ¿¡ ÀÇÁ¸Çϱ⠶§¹®¿¡ »ý±â´Â Çö»óÀÌ¸ç ±× ÀÀ¿ë¹üÀ§ÀÇ ÇѰ踦 Áö´Ï°Ô µÈ´Ù. º» ¿¬±¸¿¡¼´Â À̸¦ ÇØ°áÇϱâ À§ÇÑ Á¦¾ÈÀ¸·Î¼ Decision Tree¸¦ Harr-like Feature¿Í °áÇÕÇÏ´Â ±â¹ýÀ» Á¦¾ÈÇÑ´Ù. Decision Tree¸¦ »ç¿ë ÇÔÀ¸·Î¼ º¸´Ù ³ÐÀº ºÐ¾ßÀÇ ¹®Á¦¸¦ ÇØ°áÇϱâ À§ÇØ ±âÁ¸ÀÇ Decision Tree¸¦ Harr-like Feature¿¡ ÀûÇÕÇϵµ·Ï °³¼±ÇÑ HDCT¶ó°í ÇÏ´Â Harr-like Feature¸¦ È°¿ëÇÑ Decision Tree¸¦ Á¦¾ÈÇÏ¿´À¸¸ç ÀÌ°ÍÀÇ ¼º´ÉÀ» Adaboost¿Í ºñ±³ Æò°¡ÇÏ¿´´Ù. |
¿µ¹®³»¿ë (English Abstract) |
Adaboost is widely used for Haar-like feature boosting algorithm in Face Detection. It shows very effective performance on single distribution model. But when detecting front and side face images at same time, Adaboost shows it¡¯s limitation on multiple distribution data because it uses linear combination of basic classifier. This paper suggest the HDCT, modified decision tree algorithm for Haar-like features. We still tested the performance of HDCT compared with Adaboost on multiple distributed image recognition. |
Å°¿öµå(Keyword) |
Adaboost
Haar-like
Decision Tree
Pattern Recognition
¾Æ´ÙºÎ½ºÆ®
Çϸ£-¶óÀÌÅ©
°áÁ¤Æ®¸®
ÆÐÅÏÀνÄ
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|