Edge Perpendicular Binary Coding for USM Sharpening Detection

Abstract

Unsharp masking (USM) sharpening is a basic technique for image manipulation and editing. In recent years, the detection of USM sharpening has attracted attention from image forensics point of view. After USM sharpening, overshoot artifacts, which shape image texture, are generated along image edges. By utilizing the special characteristic of the texture modification caused by the USM sharpening, a novel method called edge perpendicular binary coding is proposed in this letter to detect USM sharpening. Extensive experiments have been conducted to show the superiority of the proposed method over the existing methods.

Image Sharpening Detection Based on Difference Sets

“…As seen from (7), the larger σ is, the more edge information of the object retained in the difference image of the original image and the Gaussian filtered image. According to the edge information of the object in the image being changed because of the sharpening operation, Ding et al proposed two image sharpening detection algorithms based on edge detection [35], [36].…”

“…In addition to the overshoot artifacts, Ding et al did a series of research for image sharpening detection based on other texture information of the image, such as image sharpening detection based on the local binary pattern (LBP) feature [34], USM sharpening detection based on the edge perpendicular binary coding (EPBC) feature [35], and weak sharpening detection based on the edge perpendicular ternary coding (EPTC) feature [36]. Different from EPBC and EPTC algorithms, which only consider a one-dimensional rectangular region perpendicular to the edge, Gu et al proposed a USM sharpening detection based on sparse coding, which considers the block of an edge region [37].…”

A Systematic Analysis of Transform Coefficients and Block Decomposition for Texture Enhancement With Orthogonal Polynomials Model

“…Instead, the variations are distributed among the magnitude of all the ac coefficients when compared to the smooth and edge region. Thus, in this proposed work, the function F 2 is defined so that it contributes to texture with the transform coefficients as given in (10). According to the functions defined above, the transformed block β under consideration is categorized as smooth, edge, and texture as…”

“…However, there is a great demand for improving visual quality. Although a number of research papers present many unswerving techniques for image enhancement, such as noise removal techniques [2]- [4], contrast improvement techniques [5], [6] and unsharp masking techniques [7]- [10], the enhancement of particular characteristics by keeping the remaining features unaltered or suppressed in the spatial and frequency domain has been a key research area in object recognition and computer vision applications.…”

“…The well-known algorithms based on histogram equalization [11]- [13] have gained popularity because they distribute the histogram of an image uniformly. Other popular unsharp masking [7]- [10] techniques act as a sharpening filter by highlighting only its high-frequency counterparts. However, noise and artifacts in the image are enhanced as well.…”