Comparison between WLD and LBP descriptors for non-intrusive image forgery detection

Abstract

Abstract-Due to the availability of easy-to-use and powerful image editing tools, the authentication of digital images cannot be taken for granted and it gives rise to non-intrusive forgery detection problem because all imaging devices do not embed watermark. We investigated the detection of copy-move and splicing, the two harmful types of image forgery, using textural properties of images. Tampering distorts the texture micropatterns in an image and texture descriptors can be employed to detect tampering. We did comparative study to examine the effect of two state-of-the-art best texture descriptors: Multiscale Local Binary Pattern (Multi-LBP) and Multiscale Weber Law Descriptor (Multi-WLD). Multiscale texture descriptors extracted from the chrominance components of an image are passed to Support Vector Machine (SVM) to identify it as authentic or forged. The performance comparison reveals that Multi-WLD performs better than Multi-LBP in detecting copymove and splicing forgeries. Multi-WLD also outperforms stateof-the-art passive forgery detection techniques.

Detecting Splicing and Copy-Move Attacks in Color Images

“…Table III shows that our proposed method produces 1.83% ~ 1.11% lower FNR than the existing method in [12]. In addition, our method performs better in terms of [12] 96.69 96.97 97.49 Muhammad et al [27] 96.39 94.89 97.33 Hussain et al [25] 94.29 --Zhao et al [11] 85.00 94.70 -He et al [ implemented by [12] Not all works have experimented with all three datasets mentioned above and therefore, here we only report results on the specific dataset(s) they reported. Table IV shows that our method outperforms existing state-of-the-art methods in all three benchmark datasets.…”

“…Some researchers utilized texture descriptors like Weber Local Descriptor (WLD) and LBP to model image tampering artifacts. In [25], Hussain et al compared multiscale WLD with multiscale LBP. They achieved better result using WLD (94.29% for splicing, 90.97% for copy-move) than using LBP (90.48% splicing and 85.83% for copy-move) on CASIA 1 dataset.…”

Review on Local Binary Patterns Variants as Texture Descriptors for Copy-Move Forgery Detection

“…Thus, a Gaussian low-pass filter is used in pre-processing to improve image quality by removing noise contained therein where filtering by more than twice can increase the detection performances [23]. The properties of LBP is capable of reducing the computational complexity problem [24]. The matching process is done by calculating the Euclidean distances for each block of the image.…”

Enhanced Fiilltterr-based SIFT Apprroach fforr Copy-Move Forrgerry Dettecttiion

“…utilized to detect the tampering by discovering distortion in the texture patterns of an image and texture descriptors [12]. The most efficient algorithms to defeat copy move forgery are invariant key-points algorithms that depend on a large number of local features extracted from an image.…”