An Iterative Approach to Super-Resolution using Multiple Low-Resolution Images

Keeping in view the increasing requirements of super resolution in a wide range of applications, this paper adds to the existing literature of alternative algorithmic approach to develop a high resolution image from multiple low resolution images. The paper, after a review of the existing methods in super resolution, presents an iterative approach using multiple low-resolution images building on top of established tools like New Edge Directed Interpolation, through a novel approach. The approach tries to discover higher frequencies from a multiplicity of data samples available, and preserves these edges across iterations. The proposed method has been tested on images of a range of complexities, including on satellite images, and the results are promising.

[1]  Hyun Wook Park,et al.  A high-resolution image reconstuction method from low-resolution image sequence , 2009, 2009 16th IEEE International Conference on Image Processing (ICIP).

[2]  Wei Wang,et al.  Super-Resolution Reconstruction of High-Resolution Satellite ZY-3 TLC Images , 2017, Sensors.

[3]  Hong Chang,et al.  Super-resolution through neighbor embedding , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[4]  Colin B. Clement,et al.  Image registration and super resolution from first principles , 2018, 1809.05583.

[5]  Roger Y. Tsai,et al.  Multiframe image restoration and registration , 1984 .

[6]  Peter Reinartz,et al.  Feature analysis for detecting people from remotely sensed images , 2013 .

[7]  Nidhi B. Patel,et al.  A Survey on Image Enhancement using Image Super-Resolution and Deblurring Methods , 2014 .

[8]  Moon Gi Kang,et al.  Super-resolution image reconstruction: a technical overview , 2003, IEEE Signal Process. Mag..

[9]  William T. Freeman,et al.  Example-Based Super-Resolution , 2002, IEEE Computer Graphics and Applications.

[10]  S. P. Kim,et al.  Subpixel accuracy image registration by spectrum cancellation , 1993, 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[11]  Robert L. Stevenson,et al.  Spatial Resolution Enhancement of Low-Resolution Image Sequences A Comprehensive Review with Directions for Future Research , 1998 .

[12]  Truong Quang Vinh,et al.  Image Super-Resolution Using Image Registration and Neural Network Based Interpolation , 2016, 2016 International Conference on Advanced Computing and Applications (ACOMP).

[13]  Zhongyuan Wang,et al.  Video Satellite Imagery Super Resolution via Convolutional Neural Networks , 2017, IEEE Geoscience and Remote Sensing Letters.