國立虎尾科技大學 |

CNN-Based Single Image Super-Resolution Network and Biomedical Image Applications.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	CNN-Based Single Image Super-Resolution Network and Biomedical Image Applications./
作者:	Bayram, Samet.
面頁冊數:	1 online resource (55 pages)
附註:	Source: Masters Abstracts International, Volume: 57-05.
Contained By:	Masters Abstracts International57-05(E).
標題:	Electrical engineering. -
電子資源:	click for full text (PQDT)
ISBN:	9780355762617

CNN-Based Single Image Super-Resolution Network and Biomedical Image Applications.
Bayram, Samet.

CNN-Based Single Image Super-Resolution Network and Biomedical Image Applications. - 1 online resource (55 pages)

Source: Masters Abstracts International, Volume: 57-05.

Thesis (M.S.)--University of Delaware, 2018.

Includes bibliographical references

In this thesis, we propose a convolutional neural network (CNN) based single image super-resolution network model with sparse representation by combining three existing state-of-the-art methods SC, SRCNN and SCN models with a modified pre-processing step. Firstly, standard gaussian box filter is applied to test image, which is a low-resolution image (LR), to remove low-frequency noises. After that, the given low-resolution image is up-scaled by bicubic interpolation method to the same size with desired output high-resolution image (HR). Secondly, a convolutional neural network based dictionary learning method is employed to train input low-resolution image to obtain LR image patches. Also, a rectified linear unit (ReLU) thresholds the output of the CNN to get a better LR image dictionary. Thirdly, to get optimal sparse parameters, we adopted Learned Iterative Shrinkage and Thresholding Algorithm (LISTA) network to train LR image patches. LISTA is a sparse-based network that generates sparse coefficients from each LR image patches. Finally, in the reconstruction step, corresponding high-resolution image patches are obtained by multiplying low-resolution image patches with optimal sparse coefficients. Then corresponding high-resolution image patches are combined to get final HR image. The experimental results show that our proposed method demonstrates outstanding performance compare to other state-of-the-art. The proposed method generates clear and better-detailed output high-resolution images since it is important in real life applications. The advantage of the proposed method is to combine convolutional neural network based dictionary learning and sparse-based network training with better pre-processing to create efficient and flexible single-image-super-resolution network.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9780355762617Subjects--Topical Terms:

596380
Electrical engineering.
Index Terms--Genre/Form:

554714
Electronic books.

CNN-Based Single Image Super-Resolution Network and Biomedical Image Applications.
LDR:03058ntm a2200349Ki 4500 001 919629
005 20181129115239.5
006 m o u
007 cr mn||||a|a||
008 190606s2018 xx obm 000 0 eng d
020 $a 9780355762617
035 $a (MiAaPQ)AAI10745822
035 $a (MiAaPQ)udel:13225
035 $a AAI10745822
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Bayram, Samet. $3 1194245
245 1 0 $a CNN-Based Single Image Super-Resolution Network and Biomedical Image Applications.
264 0 $c 2018
300 $a 1 online resource (55 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 57-05.
500 $a Advisers: Kenneth Eugene Barner; Mark Mirotznik.
502 $a Thesis (M.S.)--University of Delaware, 2018.
504 $a Includes bibliographical references
520 $a In this thesis, we propose a convolutional neural network (CNN) based single image super-resolution network model with sparse representation by combining three existing state-of-the-art methods SC, SRCNN and SCN models with a modified pre-processing step. Firstly, standard gaussian box filter is applied to test image, which is a low-resolution image (LR), to remove low-frequency noises. After that, the given low-resolution image is up-scaled by bicubic interpolation method to the same size with desired output high-resolution image (HR). Secondly, a convolutional neural network based dictionary learning method is employed to train input low-resolution image to obtain LR image patches. Also, a rectified linear unit (ReLU) thresholds the output of the CNN to get a better LR image dictionary. Thirdly, to get optimal sparse parameters, we adopted Learned Iterative Shrinkage and Thresholding Algorithm (LISTA) network to train LR image patches. LISTA is a sparse-based network that generates sparse coefficients from each LR image patches. Finally, in the reconstruction step, corresponding high-resolution image patches are obtained by multiplying low-resolution image patches with optimal sparse coefficients. Then corresponding high-resolution image patches are combined to get final HR image. The experimental results show that our proposed method demonstrates outstanding performance compare to other state-of-the-art. The proposed method generates clear and better-detailed output high-resolution images since it is important in real life applications. The advantage of the proposed method is to combine convolutional neural network based dictionary learning and sparse-based network training with better pre-processing to create efficient and flexible single-image-super-resolution network.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Electrical engineering. $3 596380
650 4 $a Medical imaging. $3 1180167
650 4 $a Artificial intelligence. $3 559380
655 7 $a Electronic books. $2 local $3 554714
690 $a 0544
690 $a 0574
690 $a 0800
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a University of Delaware. $b Department of Electrical and Computer Engineering. $3 1178922
773 0 $t Masters Abstracts International $g 57-05(E).
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10745822 $z click for full text (PQDT)