Keypoint-based enhanced image quality assessment

Ming Zeng, Jian Liu, Youfu Li, Qinghao Meng, Ting Yang, Zhengbiao Bai

    Research output: Chapters, Conference Papers, Creative and Literary WorksRGC 32 - Refereed conference paper (with host publication)peer-review

    1 Citation (Scopus)

    Abstract

    A simple objective quality assessment for enhanced images is proposed. It requires only scale-space keypoint information. Generally, the distribution of keypoints characterizes structural complexity of local regions. The process of quantitative evaluation involves four steps: The original image is first segmented into two parts (i.e. target region and background). Then, the key points in the original image and the enhanced image are extracted using the strategy of extrema in difference of Gaussian (DoG) scale space. The summed weights (SWs) of keypoints within the target region and background are computed respectively. Finally, the objective assessment result can be achieved by comparing the changes of SWs in two separate regions. The experimental results demonstrate that this proposed algorithm is feasible and effective. © 2011 Springer-Verlag.
    Original languageEnglish
    Title of host publicationAdvances in Computer Science and Education Application
    Pages420-427
    Volume202 CCIS
    DOIs
    Publication statusPublished - 2011
    Event2011 International Conference on Computer Science and Education, CSE 2011 - Qingdao, China
    Duration: 9 Jul 201110 Jul 2011

    Publication series

    NameCommunications in Computer and Information Science
    Volume202 CCIS
    ISSN (Print)1865-0929

    Conference

    Conference2011 International Conference on Computer Science and Education, CSE 2011
    PlaceChina
    CityQingdao
    Period9/07/1110/07/11

    Research Keywords

    • image enhancement
    • image quality assessment
    • keypoint detection

    Fingerprint

    Dive into the research topics of 'Keypoint-based enhanced image quality assessment'. Together they form a unique fingerprint.

    Cite this