Automatic Technology |
|
|
|
|
Super-resolution reconstruction for three-dimensional core CT image |
ZHANG Ting-rong, TENG Qi-zhi, LI Zheng-ji, QING Lin-bo, HE Xiao-hai |
College of Electronics and Information Engineering, Sichuan University, Chengdu 610065, China |
|
|
Abstract A + (adjusted anchored neighborhood regression) algorithm was extended to three-dimensional (3D) image super-resolution reconstruction in order to improve the resolution of three-dimensional image of core. A 3D high frequency correction A + algorithm was proposed. The existing high resolution (HR) core 3D CT image and the high frequency correction information were used to train the high and low resolution dictionary, the high frequency correction dictionary, the mapping matrix and the high frequency correction mapping matrix. In reconstruction, the matched dictionary atom and mapping matrix were searched for each input of the 3D low-resolution (LR) feature block and mapped directly to the HR space via multiplying the mapping vectors by LR feature vectors, respectively. Several 3D core CT images were experimented. The algorithm was compared with other three-dimensional super-resolution algorithms. The experimental results show that the proposed algorithm can obtain higher peak signal to noise ratio and structural similarity.
|
Received: 09 May 2017
Published: 26 June 2018
|
|
岩心三维CT图像超分辨率重建
为了提高岩心三维图像分辨率,将调整的锚点邻域回归算法(A+)扩展为三维图像超分辨率重建,提出三维高频修正A+算法.该算法利用已有的高分辨率(HR)岩心三维CT图像和高频修正信息训练高低分辨率字典、高频修正字典、映射矩阵和高频修正映射矩阵.重建时,对每个输入的三维低分辨率(LR)特征块搜索匹配的字典原子以及相应的映射矩阵和高频修正矩阵,通过LR特征向量分别与映射矩阵和高频映射矩阵相乘,直接将三维LR特征映射到HR空间.针对多组岩心三维CT图像进行实验,与其他三维超分辨率算法进行比较.实验结果表明,该算法具有较高的峰值信噪比和结构相似度.
|
|
[1] PELED S, YESHURUN Y. Super resolution in MRI:application to human white matter fiber tract visualization by diffusion tensor imaging[J]. Magnetic Resonance in Medicine, 2001, 45(1):29-35.
[2] GREENSPAN H, OZ G, KIRYATI N, et al. MRI inter-slice reconstruction using super-resolution[J]. Magnetic Resonance Imaging, 2002, 20(5):437-446.
[3] KORNPROBST P, PEETERS R, NIKOLOVA M, et al. A super resolution framework for fMRI sequences and its impact on resulting activation maps[C]//International Conference of Medical Image Computing and Computer-Assisted Intervention. Montréal:Springer, 2003:117-125.
[4] MANJÓN J V, COUPÉ P, BUADES A, et al. Non-local MRI upsampling[J]. Medical Image Analysis, 2010, 14(6):784-792.
[5] IWAMOTO Y, HAN X H, SASATANI S, et al. Super-resolution of MR volumetric images using sparse representation and self-similarity[C]//International Conference on Pattern Recognition.Tsukuba,Japan:IAPR, 2012:3758-3761.
[6] TIMOFTE R, DE V, GOOL L V. Anchored neighborhood regression for fast example-based super-resolution[C]//IEEE International Conference on Computer Vision. Sydney:IEEE, 2013:1920-1927.
[7] TIMOFTE R, SMET V D, GOOL L V. A+:adjusted anchored neighborhood regression for fast super-resolution[J]. Lecture Notes in Computer Science, 2015, 9006:111-126.
[8] ZEYDE R, ELAD M, PROTTER M. On single image scale-up using sparse-representations[C]//International Conference on Curves and Surfaces. Avignon:Springer, 2010:711-730.
[9] YANG J, WRIGHT J, HUANG T S, et al. Image super-resolution via sparse representation[J] IEEE Transactions on Image Processing, 2010, 19(11):2861-2873.
[10] WANG Z, BOVIK A C, SHEIKH H R, et al. Image quality assessment:from error visibility to structural similarity[J]. IEEE Transactions on Image Processing, 2004, 13(4):600-612. |
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
|
Shared |
|
|
|
|
|
Discussed |
|
|
|
|