Title:
HIGHLIGHT AREA SELF-ADAPTIVE DODGING METHOD FOR MIRROR-LIKE OBJECT
Document Type and Number:
WIPO Patent Application WO/2019/114453
Kind Code:
A1
Abstract:
Provided in the present invention is a highlight area self-adaptive dodging method for a mirror-like object. The method comprises: carrying out two-layer wavelet decomposition and reconstruction on a collected image to extract a highlight area of the image; and carrying out a local search image repairing algorithm to realize self-adaptive dodging processing of the highlight area. The method can eliminate the affects of the highlight area on defect detection in a detection process, does not need manual interferences and can automatically perform dodging on the highlight area information according to the information of a source image. Compared with other dodging algorithms, the present method can obtain a more uniform processing result and has better adaptability.
Inventors:
DU JUAN (CN)
CHEN FANG (CN)
HU YUEMING (CN)
CHEN FANG (CN)
HU YUEMING (CN)
Application Number:
PCT/CN2018/113245
Publication Date:
June 20, 2019
Filing Date:
October 31, 2018
Export Citation:
Assignee:
UNIV SOUTH CHINA TECH (CN)
International Classes:
G06T7/00
Domestic Patent References:
| WO2016176209A1 | 2016-11-03 |
Foreign References:
| CN108109138A | 2018-06-01 | |||
| CN106803240A | 2017-06-06 | |||
| CN105354802A | 2016-02-24 |
Other References:
ZHANG, ZHEN ET AL.: "Improved MASK Dodging method based on Wavelet", JOURNAL OF REMOTE SENSING, vol. 13, no. 6, 15 November 2009 (2009-11-15), pages 1079, XP055618722
LIU, CHUN: "The Research on fast image inpainting algorithms", ELECTR. TECHN. & INF. SCIENCE, CHINA MASTER'S THESES, no. 7, 15 July 2013 (2013-07-15), pages 30 - 36
LIU, CHUN: "The Research on fast image inpainting algorithms", ELECTR. TECHN. & INF. SCIENCE, CHINA MASTER'S THESES, no. 7, 15 July 2013 (2013-07-15), pages 30 - 36
Attorney, Agent or Firm:
YOGO PATENT & TRADEMARK AGENCY LIMITED COMPANY (CN)
Download PDF: