2014³â ÄÄÇ»ÅÍÁ¾ÇÕÇмú´ëȸ
Current Result Document : 183 / 184
ÇѱÛÁ¦¸ñ(Korean Title) |
iOS ±â¹ÝÀÇ Çâ»óµÈ Â÷·® ¹øÈ£ÆÇ °ËÃâÀ» À§ÇÑ 2´Ü°è ÇÕ¼º°ö ½Å°æ¸Á Á¢±Ù¹ý |
¿µ¹®Á¦¸ñ(English Title) |
Two-Step Convolutional Neural Network Approach for Improved Number Plate Localization on iOS |
ÀúÀÚ(Author) |
Å©¸®½ºÂù °Å¹ö
Á¤¸ñµ¿
Christian Gerber
Mokdong Chung
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 41 NO. 01 PP. 0868 ~ 0870 (2014. 06) |
Çѱ۳»¿ë (Korean Abstract) |
|
¿µ¹®³»¿ë (English Abstract) |
A method is proposed to achieve an improved number plate localization on iOS by applying a two-step convolutional neural network (CNN) approach. Car detection, based on a supervised CNN-verifier, is processed in the first step. In the second step, we apply the detected car image regions to the second supervised CNN-verifier for license plate detection. Since mobile devices are limited in computation power, we propose a fast method to detect number plates with a high detection rate for mobile devices. The expected areas to be used, is within the Intelligent Transportation Systems (ITS).
|
Å°¿öµå(Keyword) |
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|