Detecting Image Sharpening Using Gradient Based Directional Difference Method

Image sharpening is a widely used image enhancement operation intended to improve visual clarity by emphasizing edges and fine details. Although sharpening enhances perceptual quality, it introduces specific statistical artifacts that alter the natural relationships among neighboring pixels. Detecting such artifacts is an important task in digital image forensics, particularly for verifying image authenticity. This paper presents a difference-set based approach for detecting image sharpening operations by analyzing pixel intensity variations across multiple directions. The proposed method computes first-order and second-order difference sets over the entire image to capture global and local changes caused by sharpening. Unlike traditional edge-based methods, the proposed framework does not rely solely on edge detection and instead exploits pixel relationship statistics from all image regions. Experimental analysis demonstrates that the proposed method effectively distinguishes sharpened images from original ones under different sharpening strengths. The simplicity and robustness of the approach make it suitable for practical forensic applications.