Face spoofing is considered to be one of the prominent threats to face recognition systems. However, in order to improve the security measures of such biometric systems against deliberate spoof attacks, liveness detection has received significant recent attention from researchers. For this purpose, analysis of facial skin texture properties becomes more popular because of its limited resource requirement and lower processing cost. The traditional method of skin analysis for liveness detection was to use Local Binary Pattern (LBP) and its variants. LBP descriptors are effective, but they may exhibit certain limitations in near uniform patterns. Thus, in this paper, we demonstrate the effectiveness of Local Ternary Pattern (LTP) as an alternative to LBP. In addition, we adopted Dynamic Local Ternary Pattern (DLTP), which eliminates the manual threshold setting in LTP by using Weber’s law. The proposed method was tested rigorously on four facial spoof databases: three are public domain databases and the other is the Universiti Putra Malaysia (UPM) face spoof database, which was compiled through this study. The results obtained from the proposed DLTP texture descriptor attained optimum accuracy and clearly outperformed the reported LBP and LTP texture descriptors.
This paper proposes a robust digital image watermarking scheme that integrates block-based Singular Value Decomposition (SVD) with a three- level Discrete Wavelet Transform (DWT) preprocessing stage applied to the watermark. The host image is partitioned into non- overlapping 16×16 pixel blocks, with SVD applied independently to each block, while the watermark is decomposed through three successive Haar wavelet levels to yield a compact 32×32 approximation subband (cA3) whose 1024 elements align precisely with the 1024 image blocks in a lossless one- to- one mapping, eliminating the resizing- induced information loss characteristic of conventional hybrid schemes. Each watermark coefficients is embedded additively into the dominant singular value of its corresponding block, and extraction is accomplished in a semi- blind fashion using the stored SVD components and DWT detail subbands as secret keys. Tests on standard benchmark images show that the proposed scheme yields an average PSNR of roughly 50 dB and an SSIM close to 1, indicating that the embedded watermark is practically invisible. When subjected to eight common signal- processing attacks, the scheme reaches a normalized correlation of 0.9657 after JPEG compression, which exceeds the compression robustness reported for earlier SVD- based and hybrid DWT-SVD watermarking methods, and it remains competitive under filtering and noise attacks. These results support its use for copyright protection of digital images.