Effects of the deep learning-based super-resolution method on thermal image classification applications

被引:4
|
作者
Senalp, Fatih Mehmet [1 ]
Ceylan, Murat [1 ]
机构
[1] Konya Tech Univ, Dept Elect Elect Engn, Konya, Turkey
关键词
Thermal imaging; Super-resolution; Deep learning; Data sets; Classification;
D O I
10.1007/s11042-021-11436-4
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Thermal imaging can be used in many sectors such as public security, health, and defense in image processing. However, thermal imaging systems are very costly, limiting their use, especially in the medical field. Also, thermal camera systems obtain blurry images with low levels of detail. Therefore, the need to improve their resolution has arisen. Here, super-resolution techniques can be a solution. Developments in deep learning in recent years have increased the success of super-resolution (SR) applications. This study proposes a new deep learning-based approach TSRGAN model for SR applications performed on a new dataset consisting of thermal images of premature babies. This dataset was created by downscaling the thermal images (ground truth) of premature babies as traditional SR studies. Thus, a dataset consisting of high-resolution (HR) and low-resolution (LR) thermal images were obtained. SR images created due to the applications were compared with LR, bicubic interpolation images, and obtained SR images using state-of-the-art models. The success of the results was evaluated using image quality metrics of peak signal to noise ratio (PSNR) and structural similarity index measure (SSIM). The results show that the proposed model achieved the second-best PSNR value and the best SSIM value. Additionally, a CNN-based classifier model was developed to perform task-based evaluation, and classification applications were carried out separately on LR, HR, and reconstructed SR image sets. Here, the success of classifying unhealthy and healthy babies was compared. This study showed that the classification accuracy of SR images increased by approximately 5% compared to the classification accuracy of LR images. In addition, the classification accuracy of SR thermal images approached the classification accuracy of HR thermal images by about 2%. Therefore, with the approach proposed in this study, it has been proven that LR thermal images can be used in classification applications by increasing their resolution. Thus, widespread use of thermal imaging systems with lower costs in the medical field will be achieved.
引用
收藏
页码:9313 / 9330
页数:18
相关论文
共 50 条
  • [31] A SPATIAL CONSTRAINT AND DEEP LEARNING BASED HYPERSPECTRAL IMAGE SUPER-RESOLUTION METHOD
    Hu, Jing
    Li, Yunsong
    Zhao, Xi
    Xie, Weiying
    2017 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS), 2017, : 5129 - 5132
  • [32] Deep learning-based super-resolution in coherent imaging systems
    Liu, Tairan
    de Haan, Kevin
    Rivenson, Yair
    Wei, Zhensong
    Zeng, Xin
    Zhang, Yibo
    Ozcan, Aydogan
    SCIENTIFIC REPORTS, 2019, 9 (1)
  • [33] Volumetric Isosurface Rendering with Deep Learning-Based Super-Resolution
    Weiss, Sebastian
    Chu, Mengyu
    Thuerey, Nils
    Westermann, Rudiger
    IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS, 2021, 27 (06) : 3064 - 3078
  • [34] Deep learning-based super-resolution in coherent imaging systems
    Tairan Liu
    Kevin de Haan
    Yair Rivenson
    Zhensong Wei
    Xin Zeng
    Yibo Zhang
    Aydogan Ozcan
    Scientific Reports, 9
  • [35] Dictionary learning-based image super-resolution for multimedia devices
    Patel, Rutul
    Thakar, Vishvjit
    Joshi, Rutvij
    MULTIMEDIA TOOLS AND APPLICATIONS, 2023, 82 (11) : 17243 - 17262
  • [36] Dictionary learning-based image super-resolution for multimedia devices
    Rutul Patel
    Vishvjit Thakar
    Rutvij Joshi
    Multimedia Tools and Applications, 2023, 82 : 17243 - 17262
  • [37] Deep Learning based Frameworks for Image Super-Resolution and Noise-Resilient Super-Resolution
    Sharma, Manoj
    Chaudhury, Santanu
    Lall, Brejesh
    2017 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2017, : 744 - 751
  • [38] Image Super-Resolution Method Based on Dual Learning
    Qiu, Zhao
    Zhuang, Chunyu
    Liu, Lihao
    Lin, Jiale
    Yuan, Sheng
    JOURNAL OF CIRCUITS SYSTEMS AND COMPUTERS, 2022, 31 (16)
  • [39] Deep Learning Based Super Resolution and Classification Applications for Neonatal Thermal Images
    Senalp, Fatih M.
    Ceylan, Murat
    TRAITEMENT DU SIGNAL, 2021, 38 (05) : 1361 - 1368
  • [40] ROBUST LEARNING-BASED SUPER-RESOLUTION
    Kim, Changhyun
    Choi, Kyuha
    Lee, Ho-young
    Hwang, Kyuyoung
    Ra, Jong Beom
    2010 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, 2010, : 2017 - 2020