Source-linked AI summary
Steganography Algorithm to Hide Secret Message inside an Image
Rosziati Ibrahim, Teoh Suk Kuan
TL;DR
The paper addresses hiding data in images for privacy while preserving image quality. It proposes a bitmap-based steganography algorithm that compresses messages, converts them and a key into binary codes, and embeds those codes in pixels through SIS. A 1.0 MB BMP image reportedly hides a 3.16 KB zipped file containing 10,553 characters with near-zero visible distortion, while PSNR-based evaluation reports high-quality stego images.
Problem
Privacy concerns motivate hiding important data inside images to protect it from disclosure.
Method
The algorithm zips a text message, converts the zipped file and key into binary codes, and embeds successive two-bit groups in BMP-image pixels through SIS.
Results
3.16 KB of zipped data, containing 10,553 characters with spaces, can be hidden in a 1.0 MB BMP image with near-zero distortion; PSNR results indicate high-quality stego images.
Takeaways & Limitations
The proposed algorithm and SIS provide image-based data hiding and retrieval while keeping visible distortion near zero in the reported test.
Takeaways & Limitations
The cover image must be at least 150 pixels wide and 112 pixels high for data hiding.
Abstract
from arXiv · showhide
In this paper, the authors propose a new algorithm to hide data inside image using steganography technique. The proposed algorithm uses binary codes and pixels inside an image. The zipped file is used before it is converted to binary codes to maximize the storage of data inside the image. By applying the proposed algorithm, a system called Steganography Imaging System (SIS) is developed. The system is then tested to see the viability of the proposed algorithm. Various sizes of data are stored inside the images and the PSNR (Peak signal-to-noise ratio) is also captured for each of the images tested. Based on the PSNR value of each images, the stego image has a higher PSNR value. Hence this new steganography algorithm is very efficient to hide the data inside the image.
1. Introduction
The paper motivates hiding data inside images to protect privacy as connected systems increase disclosure risks. It proposes and implements a steganography system for embedding and retrieving data, with the paper evaluating image quality using PSNR.
- The proposed algorithm hides input data inside images, allowing recipients with the same system to retrieve and read the concealed content.
- Steganography Imaging System (SIS) implements the proposed image-based data-hiding approach.
- Growing connectivity and privacy concerns motivate hiding sensitive data inside images to reduce unauthorized disclosure.
- The paper tests different data sizes and evaluates stego-image quality using PSNR.
2. Related Work
Related work places the study within data hiding, steganography, and image-based information concealment. The authors adapt a bitmap-pixel approach that combines embedding with a later retrieval stage.
- Steganography hides the existence of a message by embedding information into a carrier.
- Prior image steganography includes methods such as least significant bits, bit-plane complexity segmentation, and chaos-based spread spectrum techniques.
- Earlier work explored hiding data in audiovisual files and embedding large image, audio, or text files inside bitmap images.
- This research uses bitmap pixels to embed data and includes separate stages for hiding and decrypting or retrieving the concealed content.
3. Proposed Algorithm
The proposed algorithm uses two security layers, compressed text and key data converted to binary codes, and pixel-level embedding with a corresponding extraction process. A secret key verifies retrieval while distributing two bits across each encoded pixel to limit image changes.
- The algorithm uses two layers of security for hiding and retrieving secret messages.
- The system requires a username and password before users can embed data and a secret key to retrieve the hidden message.
- Embedding transfers the secret message to a text file, compresses it, converts the zipped file and key into binary codes, and encodes successive two-bit groups into image pixels.The paper states that each final two-bit group is encoded into one pixel and repeated until all binary codes are embedded.
- Encoding only the last two bits per pixel is intended to limit changes to the original image.
- Extraction reconstructs binary data into a zipped text file, unzips it, and transfers the recovered content to retrieve the original message.
- The secret key is embedded with the data and used to verify that the retrieved message is correct.
4. Result and Discussion
The Steganography Imaging System implements the proposed hiding algorithm with image and message inputs, secret-key security, and retrieval support. Tests report near-zero visible distortion, high PSNR-based image quality, and substantial message capacity in BMP images.
- Algorithm and implementation: SIS provides image and text inputs, embeds a secret key with the data, and supports sending and later retrieving the hidden message.The interface requires a six-character key, which is entered twice for verification.
- Capacity and image quality: The stego images showed almost zero visible distortion compared with cover images, including the examples in Figures 6 and 7.The reported visual comparison found no noticeable distortion to the naked eye, although the stego image size was slightly higher.
- Capacity and image quality: The hiding process requires cover images at least 150 pixels wide and 112 pixels high.This is stated as the minimum pixel requirement for the data-hiding process.
- Capacity and image quality: A 1.0 MB BMP image stored a 3.16 KB zipped file, equivalent to 10553 characters with spaces or 1508 words, with near-zero distortion.The authors describe this capacity as approximately four pages of words.
- Algorithm and implementation: The system uses BMP images, zips the data to reduce file size, and encodes binary codes into image pixels.BMP is selected because its relatively large pixels provide more space for binary-code encoding.
5. Conclusions
The paper presents the Steganography Imaging System (SIS), implementing a new algorithm to hide data in images. Tests with varied payload sizes found no noticeable visual distortion and higher PSNR for stego images.
- SIS implements the proposed steganography algorithm as a system for hiding data inside images.The system was developed with two layers of security.
- Tests used images and hidden data of various sizes to evaluate the proposed algorithm.
- The stego images showed no noticeable distortion when inspected with the naked eye.
- Higher PSNR values were obtained for the stego images, indicating higher image quality under the paper’s stated evaluation.
- The paper concludes that the algorithm efficiently hides data inside images while SIS supports privacy, confidentiality, and accuracy.