Abstract : The paper presents a metric for visual security evaluation of encrypted images based on object recognition using the Scale Invariant Feature Transform (SIFT). The metrics’ behavior is demonstrated using three different encryption methods and its performance is compared to that of the PSNR, SSIM and Local Feature Based Visual Security Metric (LFBVSM). Superior correspondance to human perception and better responsiveness to subtle changes in visual security are observed for the new metric.
https://hal.inria.fr/hal-01391369 Contributor : Hal IfipConnect in order to contact the contributor Submitted on : Thursday, November 3, 2016 - 11:11:19 AM Last modification on : Thursday, March 5, 2020 - 5:41:18 PM Long-term archiving on: : Saturday, February 4, 2017 - 1:28:32 PM
Stefan Jenisch, Andreas Uhl. Visual Security Evaluation Based on SIFT Object Recognition. 10th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI), Sep 2014, Rhodes, Greece. pp.624-633, ⟨10.1007/978-3-662-44654-6_62⟩. ⟨hal-01391369⟩