Partial sharpness index for classifying digital camera and scanned images

This paper presents a new method for distinguishing images captured using a digital camera from those captured using a flatbed scanner. The method is based on the observation that edge quality of a digital image is dependent on the image source. Conceptually, the residual noise present in a digital camera image will always reduce the quality of the captured image especially the edges while the noise present in the digital camera is lower than present in the scanned image. For that reason, the edges of the images obtained from digital cameras should be sharper than the edges of the images obtained from the scanners. Thus, a classifying method for scanned and digital camera image using partial sharpness index based on local phase coherence is proposed. The experimental results show that the classification performance of our proposed method is 95%, on average, and is superior to the previous method.