Browsing by Author "Imtiaz MA"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Color Image Encryption Algorithm Based on Hyper-Chaos and DNA Computing
    (IEEE, 2020-04-24) Malik MGA; Bashir Z; Iqbal N; Imtiaz MA; Gambino O
    In this study, a novel technique using a hyper chaotic dynamical system and DNA computing has been designed with high plaintext sensitivity. In order to reduce cost, a selection procedure using tent map has been employed for generating different key streams from the same chaotic data obtained from the iterations of chaotic dynamical system. After separating the three channels from the input color image, they are both confused and diffused. First of all, these channels are diffused on a decimal level. Then they are permuted. Further, DNA encoding is performed upon these channels. Moreover, DNA level diffusion is performed to further increase the degree of randomness in the image. Lastly, the DNA encoded image is converted into decimal to get the final cipher image. Both the experimental results and security analysis strongly demonstrate the robustness of the proposed scheme. A comparison of the proposed scheme has also been made with other recently developed schemes to show that this scheme outperforms the others in terms of computational cost, time and memory efficiency. Additionally, with the large key space, the proposed scheme can resist any brute force, plaintext and statistical attacks, therefore it is a good fit for the real world applications of the image security.
  • Thumbnail Image
    Item
    Towards an Explainable Machine Learning Framework for Sketched Diagram Recognition
    (CEUR-WS Team, 2023-01-01) Singh A; Imtiaz MA; Blagojevic R; Smith-Renner A; Taele P
    In recent years, machine learning has made significant advancements in various fields, including image recognition. However, the complexity of these models often makes it difficult for users to understand the reasoning behind their predictions. This is especially true for sketch recognition, where the ability to understand and explain the model's decision-making process is crucial. To address this issue, our research focuses on developing an explainable machine learning framework for sketch recognition. The framework incorporates techniques such as feature visualization and feature attribution methods which provide insights into the model's decision-making process. The goal of this research is to not only improve the performance of sketch recognition models but also to increase their interpretability, making them more usable and trustworthy for users.