Skip to main content

Table 2 Descriptions and sizes of each data set used for training and validation

From: Printing and scanning investigation for image counter forensics

Data set name

Description

Size

Original

IFSTC data set after six manipulations

198,624

Xerox1

Images from IFSTC data set with manipulations after being printed and scanned on Xerox1

2142

Composite Printers

Combined set of images from each printer (balanced)

6426

Composite Full

Combined set of images from each printer plus original IFSTC images (balanced)

8568

Printer Identification

Printer identification and pristine images after being printed and scanned by all three printers

3213

JPEG Compression

IFSTC data set with JPEG compression (QF=80) on all images

198,624

  1. Size refers to the number of 299 × 299 or 256 × 256 image blocks in each data set, which is then split in 75% training and 25% validation. X1 and X2 refer to Xerox1 and Xerox2 printers (Sect. 3.6), respectively. The labels refer to the labels used when training and evaluating on each data set