Steganalysis for still images with LSB Steganography - Features dataset

Citation Author(s):
Julian
Miranda
Submitted by:
Julian Miranda
Last updated:
Thu, 11/21/2019 - 20:14
DOI:
10.21227/gs67-yn65
Data Format:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

This is a dataset consisting of 8 features extracted from 70,000 monochromatic still images adapted from the Genome Project Standford's database, that are labeled in two classes: LSB steganography (1) and without LSB Steganography (0). These features are Kurtosis, Skewness, Standard Deviation, Range, Median, Geometric Mean, Hjorth Mobility, and Hjorth Complexity, all extracted from the histograms of the still images, including random spatial transformations. The steganographic function embeds five types of payloads, from 0.1 to 0.5. The training dataset includes 56,000 of these pairs of labeled images (with and without LSB Steganography), with which 5,600 images conform the dataset for each payload. The testing dataset has 14,000 observations and is equally divided as the training dataset.

Instructions: 

This is a dataset consisting of 8 features extracted from 70,000 monochromatic still images adapted from the Genome Project Standford's database, that are labeled in two classes: with (1) and without (0) LSB Steganography. In the training and testing dataset, it will be found 8 columns with the following features represented as numeric quantities: Kurtosis, Skewness, Standard Deviation, Range, Median, Geometric Mean, Hjorth Mobility, and Hjorth Complexity. There is a ninth column that expresses the class of the observation, being 0 as non-steganogram and 1 as steganogram. All the features were extracted from the histograms of the still images. Reading and processing of the dataset can be done using Pandas in Python, R or Matlab.

 

The steganographic function embeds five types of payloads, from 0.1 to 0.5. The training dataset includes 56,000 of these pairs of labeled images (with and without LSB Steganography), with which 5,600 images conform the dataset for each payload. The testing dataset has 14,000 observations and is equally divided as the training dataset.

Comments

i want cover and secrt image data set

Submitted by sarah ahmed on Mon, 10/04/2021 - 11:08