Best Ratio Size of Image in Steganography using Portable Document Format with Evaluation RMSE, PSNR, and SSIM

Abstract

Steganography is a data encryption method to hide messages into files. Files that are often used are images as a place to hide data. One type of steganography that is often used is Least Significant Bit (LSB). The LSB mechanism inserts data into the last bit of each pixel in the cover file. In this paper we build application system encryption to insert Portable Document Format (PDF) files into images. Every 700 characters stored in a PDF file are inserted into the image using LSB Steganography. The images used are divided into three size classes, namely 200*200 pixels, 400*400 pixels, and 1000*1000 pixels with 100 experimental images in each class. The results of the insertion are compared with the original image using the evaluation of Root Mean Squared Error (RMSE), Structural Similarity Index (SSIM), and Peak to Signal Ratio (PSNR). The experimental results show that an image with a size of 1000*1000 shows the best results with RMSE measurements showing an average result of 7.68e-06 dB, SSIM showing an average result of 0.999999985 dB, and PSNR showing an average result of 102.2891185 dB. From the measurement results of the evaluation method, the best ratio for inserting a PDF file into an image is that every two bits of an image can accommodate one bit of a PDF file.