Super-Resolution of Low-Quality Images Based on Compressed Sensing and Sequence Information

Image super-resolution (SR) plays a significant role in Internet of Vehicles (IoV), and is widely used in many important applications, such as object recognition and vehicle identification. However, in certain cases, the quality of acquired images is low and general SR algorithms are inapplicable. Aiming at the disadvantages of the traditional image super- resolution methods which are mainly based on interpolation and example learning, this paper utilizes compressed sensing (CS) and sequence information to present a new approach of image super-resolution. In order to keep high frequency information and reduce ringing and jagged artifacts, the proposed method takes advantage of the useful information between multiple image frames and jointly trains coupled dictionaries for the low- resolution (LR) and high-resolution (HR) image block pair. By fusing atoms of LR dictionary, the atom of HR dictionary is obtained and the HR image can be recovered in terms of the learned HR dictionary. The experimental results show that the proposed algorithm has better performance in both subjective assessment and objective standards, including Entropy and Average Gradient.

[1]  Soheil Darabi,et al.  Compressive image super-resolution , 2009, 2009 Conference Record of the Forty-Third Asilomar Conference on Signals, Systems and Computers.

[2]  Michal Irani,et al.  Motion Analysis for Image Enhancement: Resolution, Occlusion, and Transparency , 1993, J. Vis. Commun. Image Represent..

[3]  Zongxu Pan,et al.  Super-Resolution Based on Compressive Sensing and Structural Self-Similarity for Remote Sensing Images , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[4]  Ting-Zhu Huang,et al.  Single image super-resolution by approximated Heaviside functions , 2015, Inf. Sci..

[5]  Wang Wei,et al.  Parameters Estimation of High Speed Targets Based on Frequency Domain Super-resolution , 2016 .

[6]  Xiaoou Tang,et al.  Learning a Deep Convolutional Network for Image Super-Resolution , 2014, ECCV.

[7]  A. Murat Tekalp,et al.  High-resolution image reconstruction from lower-resolution image sequences and space-varying image restoration , 1992, [Proceedings] ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[8]  Aggelos K. Katsaggelos,et al.  Image super-resolution from compressed sensing observations , 2015, 2015 IEEE International Conference on Image Processing (ICIP).

[9]  Kwang In Kim,et al.  Example-Based Learning for Single-Image Super-Resolution , 2008, DAGM-Symposium.

[10]  Michael Elad,et al.  On Single Image Scale-Up Using Sparse-Representations , 2010, Curves and Surfaces.

[11]  Xinbo Gao,et al.  Single Image Super-Resolution via Multiple Mixture Prior Models , 2018, IEEE Transactions on Image Processing.

[12]  Emmanuel J. Candès,et al.  Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information , 2004, IEEE Transactions on Information Theory.

[13]  Pascal Getreuer,et al.  Contour Stencils: Total Variation along Curves for Adaptive Image Interpolation , 2011, SIAM J. Imaging Sci..

[14]  Truong Q. Nguyen,et al.  Image Superresolution Using Support Vector Regression , 2007, IEEE Transactions on Image Processing.

[15]  H Stark,et al.  High-resolution image recovery from image-plane arrays, using convex projections. , 1989, Journal of the Optical Society of America. A, Optics and image science.

[16]  Lei Zhang,et al.  An edge-guided image interpolation algorithm via directional filtering and data fusion , 2006, IEEE Transactions on Image Processing.