• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹ÀÎÅͳÝÁ¤º¸ÇÐȸ ³í¹®Áö

Çѱ¹ÀÎÅͳÝÁ¤º¸ÇÐȸ ³í¹®Áö

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) ¿ÀÅäÀÎÄÚ´õ ±â¹ÝÀÇ ÀâÀ½¿¡ °­ÀÎÇÑ °èÃþÀû À̹ÌÁö ºÐ·ù ½Ã½ºÅÛ
¿µ¹®Á¦¸ñ(English Title) A Noise-Tolerant Hierarchical Image Classification System based on Autoencoder Models
ÀúÀÚ(Author) ÀÌÁ¾°ü   Jong-kwan Lee  
¿ø¹®¼ö·Ïó(Citation) VOL 22 NO. 01 PP. 0023 ~ 0030 (2021. 02)
Çѱ۳»¿ë
(Korean Abstract)
º» ³í¹®Àº ´Ù¼öÀÇ ¿ÀÅäÀÎÄÚ´õ ¸ðµ¨µéÀ» ÀÌ¿ëÇÑ ÀâÀ½¿¡ °­ÀÎÇÑ À̹ÌÁö ºÐ·ù ½Ã½ºÅÛÀ» Á¦¾ÈÇÑ´Ù. µö·¯´× ±â¼úÀÇ ¹ß´Þ·Î À̹ÌÁö ºÐ·ùÀÇ Á¤È®µµ´Â Á¡Á¡ ³ô¾ÆÁö°í ÀÖ´Ù. ÇÏÁö¸¸ ÀÔ·Â À̹ÌÁö°¡ ÀâÀ½¿¡ ÀÇÇؼ­ ¿À¿°µÈ °æ¿ì¿¡´Â À̹ÌÁö ºÐ·ù ¼º´ÉÀÌ ±Þ°ÝÈ÷ ÀúÇϵȴÙ. À̹ÌÁö¿¡ ÷°¡µÇ´Â ÀâÀ½Àº À̹ÌÁöÀÇ »ý¼º ¹× Àü¼Û °úÁ¤¿¡¼­ ÇÊ¿¬ÀûÀ¸·Î ¹ß»ýÇÒ ¼ö¹Û¿¡ ¾ø´Ù. µû¶ó¼­ ½ÇÁ¦ ȯ°æ¿¡¼­ À̹ÌÁö ºÐ·ù±â°¡ »ç¿ëµÇ±â À§Çؼ­´Â ÀâÀ½¿¡ ´ëÇÑ Ã³¸® ¹× ´ëÀÀÀÌ ¹Ýµå½Ã ÇÊ¿äÇÏ´Ù. ÇÑÆí ¿ÀÅäÀÎÄÚ´õ´Â ÀԷ°ª°ú Ãâ·Â°ªÀÌ À¯»çÇϵµ·Ï ÇнÀµÇ¾îÁö´Â Àΰø½Å°æ¸Á ¸ðµ¨ÀÌ´Ù. ÀԷµ¥ÀÌÅÍ°¡ ÇнÀµ¥ÀÌÅÍ¿Í À¯»çÇÏ´Ù¸é ¿ÀÅäÀÎÄÚ´õÀÇ Ãâ·Âµ¥ÀÌÅÍ¿Í ÀԷµ¥ÀÌÅÍ »çÀÌÀÇ ¿ÀÂ÷´Â ÀÛÀ» °ÍÀÌ´Ù. ÇÏÁö¸¸ ÀÔ·Â µ¥ÀÌÅÍ°¡ ÇнÀµ¥ÀÌÅÍ¿Í À¯»ç¼ºÀÌ ¾ø´Ù¸é ¿ÀÅäÀÎÄÚ´õÀÇ Ãâ·Âµ¥ÀÌÅÍ¿Í ÀԷµ¥ÀÌÅÍ »çÀÌÀÇ ¿ÀÂ÷´Â Ŭ °ÍÀÌ´Ù. Á¦¾ÈÇÏ´Â ½Ã½ºÅÛÀº ¿ÀÅäÀÎÄÚ´õÀÇ ÀԷµ¥ÀÌÅÍ¿Í Ãâ·Âµ¥ÀÌÅÍ »çÀÌÀÇ °ü°è¸¦ ÀÌ¿ëÇÑ´Ù. Á¦¾ÈÇÏ´Â ½Ã½ºÅÛÀÇ À̹ÌÁö ºÐ·ù ÀýÂ÷´Â 2´Ü°è·Î ±¸¼ºµÈ ´Ù. 1´Ü°è¿¡¼­ ºÐ·ù °¡´É¼ºÀÌ °¡Àå ³ôÀº Ŭ·¡½º 2°³¸¦ ¼±Á¤ÇÏ°í À̵é Ŭ·¡½ºÀÇ ºÐ·ù °¡´É¼ºÀÌ ¼­·Î À¯»çÇϸé 2´Ü°è¿¡¼­ Ãß°¡ÀûÀÎ ºÐ·ù ÀýÂ÷¸¦ °ÅÄ£´Ù. Á¦¾ÈÇÏ´Â ½Ã½ºÅÛÀÇ ¼º´É ºÐ¼®À» À§ÇØ °¡¿ì½Ã¾È ÀâÀ½À¸·Î ¿À¿°µÈ MNIST µ¥ÀÌÅͼÂÀ» ´ë»óÀ¸·Î ºÐ·ù Á¤È®µµ¸¦ ½ÇÇèÇÏ¿´´Ù. ½ÇÇè °á°ú ÀâÀ½ ȯ°æ¿¡¼­ Á¦¾ÈÇÏ´Â ½Ã½ºÅÛÀÌ CNN(Convolutional Neural Network) ±â¹ÝÀÇ ºÐ·ù ±â¹ý¿¡ ºñÇØ ³ôÀº Á¤È®µµ¸¦ ³ªÅ¸³¿À» È®ÀÎÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
This paper proposes a noise-tolerant image classification system using multiple autoencoders. The development of deep learning technology has dramatically improved the performance of image classifiers. However, if the images are contaminated by noise, the performance degrades rapidly. Noise added to the image is inevitably generated in the process of obtaining and transmitting the image. Therefore, in order to use the classifier in a real environment, we have to deal with the noise. On the other hand, the autoencoder is an artificial neural network model that is trained to have similar input and output values. If the input data is similar to the training data, the error between the input data and output data of the autoencoder will be small. However, if the input data is not similar to the training data, the error will be large. The proposed system uses the relationship between the input data and the output data of the autoencoder, and it has two phases to classify the images. In the first phase, the classes with the highest likelihood of classification are selected and subject to the procedure again in the second phase. For the performance analysis of the proposed system, classification accuracy was tested on a Gaussian noise-contaminated MNIST dataset. As a result of the experiment, it was confirmed that the proposed system in the noisy environment has higher accuracy than the CNN-based classification technique.
Å°¿öµå(Keyword) À̹ÌÁö ºÐ·ù   µö·¯´×   ¸Ó½Å·¯´×   ¿ÀÅäÀÎÄÚ´õ   ÀâÀ½   Image classification   Deep learning   Machine learning   Autoencoder   Noise  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå