Çankaya GCRIS Standart veritabanının içerik oluşturulması ve kurulumu Research Ecosystems (https://www.researchecosystems.com) tarafından devam etmektedir. Bu süreçte gördüğünüz verilerde eksikler olabilir.

Browsing by Author "Mohammed, Ahmed"

Filter results by typing the first few letters
Now showing 1 - 3 of 3
  • Thumbnail Image
    Conference Object
    Citation Count: Maras, Hakan Hadi; Mohammed, Ahmed; Elbasi, Ersin, "A New Robust Binary Image Embedding Algorithm in Discrete Wavelet Domain", 8th IEEE International Conference on Application of Information and Communication Technologies (AICT), pp. 16-22, (2014).
    A New Robust Binary Image Embedding Algorithm in Discrete Wavelet Domain
    (IEEE, 2014) Mohammed, Ahmed; Maraş, H. Hakan; Elbaşı, Ersin; 34410
    Digital watermarks have recently emerged as a possible solution for protecting the copyright of digital materials, the work presented in this paper is concerned with the Discrete Wavelet Transform (DWT) based non-blind digital watermarking, and how the DWT is an efficient transform in the field of digital watermarking. In this work we used an optimum criteria that embeds four watermarks in more than one level of DWT in the same algorithm. The aim of this work is to keep the Correlation Coefficient (CC) between the original and the extracted watermark around the value of 0.9.
  • Thumbnail Image
    Master Thesis
    Citation Count: MOHAMMED, A. (2015). Anomalous network packet detection. Yayımlanmamış yüksek lisans tezi. Ankara: Çankaya Üniversitesi Fen Bilimleri Enstitüsü.
    Anomalous network packet detection
    (2015) Mohammed, Ahmed
    In the last decade, extensive research has been done to the improvement of Intrusion Detection Systems (IDS) for anomalous network packets. Two types of IDS are available. The first one is the signature-based detection system. It can detect intrusions by scanning network packets and compare them with human-generated signatures against previously observed attacks. The second type is the anomaly-based detection system, which is able to detect new attacks against observed attacks. In this thesis, anomaly-based detection systems have been used with density base clustering algorithms and techniques. DBSCAN (Density-Based Spatial Clustering and Application with Noise) and DenStream algorithms are well-known data stream clustering algorithms in data mining. DBSCAN algorithm can separate packets as normal and noisy data. The second algorithm, DenStream starts off with DBSCAN and then tries to reduce the amount of noise to be clustered. For this study, we used DARPA' 99 dataset. We worked with attacks of type R2L, U2R, DoS and Probe. The DenStream and DBSCAN algorithms have been performed with fine-tuned. Overall, the DenStream algorithm achieved higher detection results and sensitivity than the DBSCAN algorithm. After, only epsilon distance and minimum number of points parameters for neighborhood area are fine-tuned, the clustering methods can be easily applied for classifying normal and noisy data regardless of its attack type
  • Thumbnail Image
    Master Thesis
    Citation Count: MOHAMMED, A. (2014). Efficiency of discrete wavelet trasnform in digital watermarking. Yayımlanmamış yüksek lisans tezi. Ankara: Çankaya Üniversitesi Fen Bilimleri Enstitüsü.
    Efficiency of discrete wavelet trasnform in digital watermarking
    (2014) Mohammed, Ahmed
    Digital watermarks have recently emerged as a possible solution for protecting the copyright of digital materials. The work presented in this thesis is concerned with the Discrete Wavelet Transform DWT based digital watermarking, and how the DWT is an efficient transform in the field of digital watermarking. Four efficiency test algorithms were proposed in this work, each uses a different level of DWT decomposition as an embedding domain for the binary pattern watermarks used, the four efficiency test algorithms were applied on both gray scale and colored images with RGB (green and blue channels were used only) and Ycbcr (Y luminance was used only) color spaces. Fifteen important watermark attacks were applied on the four algorithms. Results were studied and examined carefully, and that led to obtain an optimum algorithm which is based on multi-level embedding criteria. The optimum algorithm was tested against even harder attacks which were the dual attacks (more than one attacks at a time) and a Hard crop attack that removes almost 90% of the watermarked image data, then the performance of the optimum algorithm has been compared with some selected literature algorithms in the field