周圆, 王爱华, 陈莹, 侯春萍
AuthorsHTML:周圆, 王爱华, 陈莹, 侯春萍
AuthorsListE:Zhou Yuan, Wang Aihua, Chen Ying, Hou Chunping
AuthorsHTMLE:Zhou Yuan, Wang Aihua, Chen Ying, Hou Chunping
Unit:天津大学电气自动化与信息工程学院，天津 300072
Unit_EngLish:School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China
Abstract_Chinese:针对在立体图像的超分辨率重建过程中, 需要分别对低分辨率的彩图和同场景的深度图进行超分辨率重建的问题, 提出了一种基于联合稀疏表示的立体图像的超分辨率重建方法．该方法在非局部中心稀疏表示重建方法的基础上, 利用彩色图像与同场景深度图像的耦合相关性, 通过构造联合特征图像块来学习彩色和深度图像的联合字典; 然后构造彩色和深度图像块的联合编码增量作为正则项, 利用迭代优化算法求解模型, 进而同时重建高分辨率的彩色和深度图像．为验证算法的有效性, 在Middlebury数据集上对重建结果进行了主、客观评估, 并与不同算法进行了比较. 实验结果表明, 在客观指标和主观视觉效果上, 本文提出的算法可以同时获得令人满意的彩图和高质量的深度图.
Abstract_English:To resolve the stereo image super-resolution reconstruction problem of reconstructing the color image and its corresponding depth map in the same scene respectively，a stereo image super-resolution reconstruction method based on the joint sparse representation is proposed in this paper. The model is developed based on the nonlocal center sparse representation method，using the correspondence between the color image and the depth map of the same scene. The joint dictionary is learned through joint characteristic image tiles construction. Then regularize the problem with joint coding increments of color and depth image tiles. Subsequently，an iterative optimization algorithm is applied to solve the proposed model. The high-resolution color image and its corresponding depth image are restored simultaneously. To evaluate the effectiveness of the proposed algorithm，several experiments on the Middlebury dataset are conducted，and the proposed algorithm and different methods in both objective indexes and subjective visual experience are compared. The experimental results show that the proposed algorithm achieves satisfactory super-resolution results in both objective indexes and subjective visual comparisons.
Keyword_Chinese:超分辨率重建; 稀疏表示; 联合特征图像块; 立体图像; 联合字典学习
Keywords_English:super-resolution reconstruction; sparse representation; joint characteristic image tiles; stereo image; joint dictionary learning

PDF全文下载地址:http://xbzrb.tju.edu.cn/#/digest?ArticleID=5809