Deep learning-based image de-raining using discrete Fourier transformation
- PDF / 8,943,164 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 16 Downloads / 251 Views
ORIGINAL ARTICLE
Deep learning-based image de-raining using discrete Fourier transformation Analyzing behaviour of deep CNNs towards uncorrelated transformed domain data Prasen Kumar Sharma1
· Sathisha Basavaraju1 · Arijit Sur1
Accepted: 31 August 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Single image rain streak removal is a well-explored topic in the field of computer vision. The de-raining problem is modeled as an image decomposition task where a rainy image is decomposed into rain-free background image and rain streek map. Unlike most of the existing de-raining methods, this paper attempts to decompose the rainy image in the frequency domain. The idea is inspired by pseudo-periodic characteristics of the noise signal (here the rain streaks) which leave some traces in the frequency domain, and the same can be utilized to predict the noise signal. In this paper, a deep learning-based rain streak prediction model is proposed which learns in discrete Fourier transform Oppenheim and Schafer (Discrete-TimeSignal Processing, Prentice Hall, Upper Saddle River, 1989) domain. To the best of our knowledge, this is the first approach where compressed domain coefficients are directly used as input to a deep convolutional neural network. The proposed model has been tested on publicly available synthetic datasets Fu et al. (in: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. https://doi.org/10.1109/CVPR.2017.186, Yang et al. (in: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. https://doi.org/10.1109/CVPR.2017.183), Yeh et al. (in: 2015 IEEE International Conference on Consumer Electronics-Taiwan, 2015. https://doi.org/10.1109/ICCE-TW.2015.7216999) and results are found to be comparable with the state of the art methods in the spatial domain. The presented analysis and study have an obvious indication to extend transform domain input to train the deep learning architecture especially image de-noising like problems. Keywords Image de-raining · Deep learning · Convolutional neural networks · Discrete Fourier transformation
1 Introduction Rain streak removal from a single image or video has been given considerable importance in recent computer vision literature. In the real world scenario, rain streaks may cause problems in many image-based applications such as surveillance, satellite-based image tracking, autonomous car driving, etc. It can also be taken as a noise removal problem from a single image or video. Rain streak removal from a sin-
B
Prasen Kumar Sharma [email protected] Sathisha Basavaraju [email protected] Arijit Sur [email protected]
1
Multimedia Laboratory, Department of Computer Science and Engineering, Indian Institute of Technology, Guwahati, India
gle image is an image decomposition problem where a rainy image is decomposed to rain-free background image and rain streak map. There have been many methods for noise or rain streak removal where the majority of them work in the spatial domain. It has b
Data Loading...
