Steganography applied in the origin claim of pictures captured by drones based on chaos

Abstract

Abstrac: In this work, steganography is implemented in photographs captured by an unmanned aerial vehicle (drone), with the purpose of adding an identifier that indicates which device they are taken from so it works for the recovery of the origin. In the system, a new technique that modifies the least significant bit (LSB) is applied, using a mathematical model to generate the chaotic orbits, one of the parts selects the RGB channel (Red, Green or Blue) where the LSB is changed and the other is implemented to calculate the random position of the sub pixel to be modified in the selected channel. In addition, a comparison between the bit to be hidden and the LSB of the pixel of the image is performed to verify if it is not necessary to modify it, which lessens the alterations in the container image. It is a tool to capture photos remotely with the Ar.Drone 2.0, with the features needed to perform an analysis that uses correlation diagrams and histograms to verify if the integrity of the message is guaranteed or if changes in the stego-image are visible to the naked eye. On the other hand, a test was done on the Baboon image to compare the robustness of the proposed system with other investigations, evaluating the correlation, contrast, energy, homogeneity, MSE, PSNR and quality index. The results generated were compared with the work of other authors concluding our system provides greater security, integrity, high sensitivity to the keys, it is not linked to a single chaotic system and can be applied to hide imperceptibly all kinds of information, in: radiographs, videos, files, official documents, and other types of containers. Resumen En este trabajo se implementó esteganografía en las fotografías capturadas por un vehículo aéreo no tripulado (dron), con la finalidad de agregar un identificador que indique desde que dispositivo fueron tomadas, y funcione para la reclamación de origen. En el sistema se aplicó una nueva técnica que modifica el bit menos significativo (LSB) usando un modelo matemático que se utiliza para generar dos órbitas caóticas, una de las cuales se usa para seleccionar el canal RGB (Rojo, Verde o Azul), donde se cambió el LSB y la otra se implementó para calcular de forma aleatoria la posición del subpixel que se modificará en el canal seleccionado; además se realizó una comparación entre el bit que se desea ocultar y el LSB del pixel de la imagen, para verificar si no es necesario modificarlo, lo cual altera menos la imagen original. Se realizaron pruebas del algoritmo en un Ar.drone 2.0, pero se puede aplicar para ocultar de forma imperceptible todo tipo de información, en: radiografías, videos, archivos, documentos oficiales, o cualquier otro tipo de contendor.