TY - GEN
T1 - Multi-Band NIR Colorization Using Structure-Aware Network
AU - Park, Min Je
AU - Lee, Ju Han
AU - Lee, Sang Ho
AU - Kim, Jong Ok
N1 - Funding Information:
This work was supported in part by the National "Versatile visible and near-infrared image fusion based on ResearchFoundationofKorea NRF) rant undedb(ythGfhigvisibilityareaselection ."JeounalofElhectronicr Korean Government (MSIT) under Grant Imaging25.1 2016):013016. ( 2020R1A4A4079705 and in part by the Ministry of Sun, Tian, et al. "Nir to rgb domain translation using Science and ICT (MSIT), South Korea, through the asymmetric cycle generative adversarial networks."IEEE InformationTechnology ResearchCenter(ITRC)Support Limmer, Matthias, and Hendrik PA Lensch. "InfraredAccess(2019):112459-1124679. Program supervised by the Institute of Information & colorization using deep convolutional neural Communications Technology Planning & Evaluation networks." 2016 15th IEEE International Conference on (IITP)derGrantITP-202u12020007I49. n---1 [7] Dong, Ziyue, Sei-ichiro Kamata, and Toby P. Breckon. "Infrared imagecolorization usinga-shapenetwork."2018 s 25th IEEE International Conference on Image Processing (ICIP). EEE, 2018. I [8] Sun, Tian, and Cheolkon Jung. " Sun, Tian, and Cheolkon Jung. "NIRImageColorizationUsingSPADEGeneratorand Grayscale Approximated Self-Reconstruction."2020 IEEE
Funding Information:
This work was supported in part by the National Research Foundation of Korea (NRF).
Publisher Copyright:
© 2021 APSIPA.
PY - 2021
Y1 - 2021
N2 - Near infrared (NIR) image can capture the scene in the low light condition without noise unlike RGB. Therefore, it has been widely used in low light vision problems and is often fused with its RGB counterpart for RGB image enhancement. However, there can be some situations that an RGB image can be hardly captured like extremely low-light condition. To cope with this problem, researches to convert NIR image to RGB have been recently conducted based on deep learning networks. These methods show a good performance relatively, but they have some limitations of performance. In this paper, we propose a deep network to colorize multi-band NIR images to RGB using our new dataset. The proposed method attempts to exploit the correlation between individual NIR band and RGB by using multi-band NIR images. It can successfully colorize the multi-band NIR images using two-branch structure and the constraint of the proportional gradient between NIR and RGB.
AB - Near infrared (NIR) image can capture the scene in the low light condition without noise unlike RGB. Therefore, it has been widely used in low light vision problems and is often fused with its RGB counterpart for RGB image enhancement. However, there can be some situations that an RGB image can be hardly captured like extremely low-light condition. To cope with this problem, researches to convert NIR image to RGB have been recently conducted based on deep learning networks. These methods show a good performance relatively, but they have some limitations of performance. In this paper, we propose a deep network to colorize multi-band NIR images to RGB using our new dataset. The proposed method attempts to exploit the correlation between individual NIR band and RGB by using multi-band NIR images. It can successfully colorize the multi-band NIR images using two-branch structure and the constraint of the proportional gradient between NIR and RGB.
UR - http://www.scopus.com/inward/record.url?scp=85126721889&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85126721889
T3 - 2021 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2021 - Proceedings
SP - 1682
EP - 1686
BT - 2021 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2021 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2021 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2021
Y2 - 14 December 2021 through 17 December 2021
ER -