Computationally efficient image restoration and super-resolution algorithms for real-time implementation

被引:2
|
作者
Sundareshan, MK [1 ]
机构
[1] Univ Arizona, Dept Elect & Comp Engn, Tucson, AZ 85721 USA
关键词
tactical images; passive millimeter-wave images; image restoration; super-resolution;
D O I
10.1117/12.477471
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Computational complexity is a major impediment to the real-time implementation of image restoration and super-resolution algorithms. Although powerful restoration algorithms have been developed within the last few years utilizing sophisticated mathematical machinery (based on statistical optimization and convex set theory), these algorithms are typically iterative in nature and require enough number of iterations to be executed to achieve desired resolution gains in order to meaningfully perform detection and recognition tasks in practice. Additionally, recent technological breakthroughs have facilitated novel sensor designs (focal plane arrays, for instance) that make it possible to capture mega-pixel imagery data at video frame rates. A major challenge in the processing of these large format images is to complete the execution of the image processing steps within the frame capture times and to keep up with the output rate of the sensor so that all data captured by the sensor can be efficiently utilized. Consequently, development of novel methods that facilitate real-time implementation of image restoration and super-resolution algorithms is of significant practical interest and will be the primary focus of this paper. The key to designing computationally efficient processing schemes lies in strategically introducing appropriate pre-processing and post-processing steps together with the super-resolution iterations in order to tailor optimized overall processing sequences for imagery data of specific formats. Three distinct methods for tailoring a pre-processing filter and integrating it with the super-resolution processing steps will be outlined in this paper. These methods consist of a Region-of-Interest (ROI) extraction scheme, a background-detail separation procedure, and a scene-derived information extraction step for implementing a set-theoretic restoration of the image that is less demanding in computation compared to the super-resolution iterations. A quantitative evaluation of the performance of these algorithms for restoring and super-resolving tactical imagery data, including in particular PMMW images, will also be presented.
引用
收藏
页码:306 / 317
页数:12
相关论文
共 50 条
  • [1] Real-time implementation of super-resolution imaging algorithm
    Hein, C
    [J]. ADVANCED SIGNAL PROCESSING ALGORITHMS, ARCHITECTURES, AND IMPLEMENTATIONS VIII, 1998, 3461 : 501 - 511
  • [2] EFFICIENT SELF-CALIBRATED CONVOLUTION FOR REAL-TIME IMAGE SUPER-RESOLUTION
    Hamida, Adnan
    Alfarraj, Motaz
    Zummo, Salam A.
    [J]. 2022 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, ICIP, 2022, : 1176 - 1180
  • [3] Search for a computationally efficient image super-resolution algorithm
    Bannore, Vivek
    Swierkowski, Leszek
    [J]. KNOWLEDGE-BASED INTELLIGENT INFORMATION AND ENGINEERING SYSTEMS: KES 2007 - WIRN 2007, PT I, PROCEEDINGS, 2007, 4692 : 510 - +
  • [4] Image restoration and super-resolution algorithms based on frequency domain and their parallel implementation
    Ma, Dong-Dong
    Li, Jin-Zong
    Zhu, Bing
    Mu, Li-Sheng
    [J]. Guangdianzi Jiguang/Journal of Optoelectronics Laser, 2009, 20 (10): : 1365 - 1370
  • [5] Hardware Implementation of Real-Time Multiple Frame Super-Resolution
    Seyid, Kerem
    Blanc, Sebastien
    Leblebici, Yusuf
    [J]. 2015 IFIP/IEEE INTERNATIONAL CONFERENCE ON VERY LARGE SCALE INTEGRATION (VLSI-SOC), 2015, : 219 - 224
  • [6] Efficient local cascading residual network for real-time single image super-resolution
    Haoran Yang
    Qingyu Dou
    Kai Liu
    Zitao Liu
    Rita Francese
    Xiaomin Yang
    [J]. Journal of Real-Time Image Processing, 2021, 18 : 1235 - 1246
  • [7] Efficient local cascading residual network for real-time single image super-resolution
    Yang, Haoran
    Dou, Qingyu
    Liu, Kai
    Liu, Zitao
    Francese, Rita
    Yang, Xiaomin
    [J]. JOURNAL OF REAL-TIME IMAGE PROCESSING, 2021, 18 (04) : 1235 - 1246
  • [8] Deep learning approaches for real-time image super-resolution
    Shamsolmoali, Pourya
    Celebi, M. Emre
    Wang, Ruili
    [J]. NEURAL COMPUTING & APPLICATIONS, 2020, 32 (18): : 14519 - 14520
  • [9] Real-time Document Image Super-Resolution by Fast Matting
    Zheng, Yun
    Kang, Xudong
    Li, Shutao
    He, Yuan
    Sun, Jun
    [J]. 2014 11TH IAPR INTERNATIONAL WORKSHOP ON DOCUMENT ANALYSIS SYSTEMS (DAS 2014), 2014, : 232 - 236
  • [10] TherISuRNet - A Computationally Efficient Thermal Image Super-Resolution Network
    Chudasama, Vishal
    Patel, Heena
    Prajapati, Kalpesh
    Upla, Kishor
    Ramachandra, Raghavendra
    Raja, Kiran
    Busch, Christoph
    [J]. 2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW 2020), 2020, : 388 - 397