TY - JOUR
T1 - Adaptive just-noticeable difference profile for image hashing
AU - Khan, Muhammad Farhan
AU - Monir, Syed Muhammad
AU - Naseem, Imran
AU - Khan, Bilal Muhammad
N1 - DBLP License: DBLP's bibliographic metadata records provided through http://dblp.org/ are distributed under a Creative Commons CC0 1.0 Universal Public Domain Dedication. Although the bibliographic metadata records are provided consistent with CC0 1.0 Dedication, the content described by the metadata records is not. Content may be subject to copyright, rights of privacy, rights of publicity and other restrictions.
PY - 2021/3
Y1 - 2021/3
N2 - A novel algorithm to extract image features related to human perception of colors is presented in this paper. The proposed profile is based on the generation of a lookup table with thresholds for just-noticeable color difference against all possible colors from RGB color gamut. The profile is further used to reduce the computational cost of image hashing. The host color image is scanned to generate the maximum magnitude difference profile. The lookup table is then used for thresholding of magnitude differences to generate the binary adaptive just-noticeable difference profile. The generated profile can be used for image watermarking, hashing and similar image processing applications. Performance of the proposed profile has been tested for image hashing. Average similarity of 0.97 for 66 tampered versions of 42 host images and discrimination of 0.138 among 79,800 different image hash pairs have been obtained while area under receiver operating characteristic curve is 0.9980.
AB - A novel algorithm to extract image features related to human perception of colors is presented in this paper. The proposed profile is based on the generation of a lookup table with thresholds for just-noticeable color difference against all possible colors from RGB color gamut. The profile is further used to reduce the computational cost of image hashing. The host color image is scanned to generate the maximum magnitude difference profile. The lookup table is then used for thresholding of magnitude differences to generate the binary adaptive just-noticeable difference profile. The generated profile can be used for image watermarking, hashing and similar image processing applications. Performance of the proposed profile has been tested for image hashing. Average similarity of 0.97 for 66 tampered versions of 42 host images and discrimination of 0.138 among 79,800 different image hash pairs have been obtained while area under receiver operating characteristic curve is 0.9980.
KW - Feature extraction
KW - Image hashing
KW - Image profile
KW - Image thresholding
KW - Just-noticeable difference
KW - Perceptual features
UR - http://www.scopus.com/inward/record.url?scp=85100064256&partnerID=8YFLogxK
U2 - 10.1016/j.compeleceng.2020.106967
DO - 10.1016/j.compeleceng.2020.106967
M3 - Article
AN - SCOPUS:85100064256
SN - 0045-7906
VL - 90
SP - 106967
JO - Computers and Electrical Engineering
JF - Computers and Electrical Engineering
M1 - 106967
ER -