
Taha Yassin Abdulqader
Research Interests|
Gender | MALE |
---|---|
Place of Work | Mosul Technical Institute |
Position | مقرر القسم |
Qualification | Master |
Speciality | Computer Science |
taha.yassin@ntu.edu.iq | |
Phone | 07737502494 |
Address | حي المأمون, Mosul, Mosul, Iraq |
Skills
Image Processing (80%)
Watermarking (85%)
Data Hiding (80%)
Machine Learning (70%)
Publications
Enhancing Image Recognition with Quaternion Neural Networks: A Novel Approach to Color Layer Integration
Apr 25, 2025Journal Journal of Image and Graphics
publisher Scopus
DOI 10.18178/joig.13.2.190-197
Issue 2301-3699 (Print); 2972-3973 (Online)
Volume 13, No. 2, 2025
Neural networks have been widely used in image recognition tasks, and this study explores a novel method— Quaternion Neural Networks (QNNs)—for enhancing performance. Using quaternion algebra, QNNs minimize the number of trainable parameters, leading to more compact models and quicker training times than Convolutional Neural Network CNNs. Therefore, color layers might potentially improve network performance by learning common parameters through input as linked values. Experiments assess learning processes by taking into account the roles of color and structure as well as stability in the presence of noisy visuals. According to the experimental results, QNNs retain an accuracy of 85% in the absence of noise, but at a noise level of σ = 0.30, accuracy dropped to 70%. Notwithstanding this, the network proved to be effective in learning structural information, exhibiting robustness against noise and disturbances in texture and color, hence confirming its suitability for wider image recognition uses. The paper establishes a proof of concept for the effectiveness of quaternion networks that will open up new avenues for research and possible uses that could outperform or supplement traditional networks.
Robust Image Watermarking for Tamper Detection and Self-Recovery Using SVD and RSA Methods
Jul 25, 2024Journal CENTRAL ASIAN JOURNAL OF MATHEMATICAL THEORY AND COMPUTER SCIENCES
Issue 03
Volume 05
To The paper proposes an Image watermarking technique for identifying and selfrecovering tampered images. The system identifies tampered images by comparing the SVD values of 4X4 blocks and average pixel intensities of 2×2 blocks. In the process of SVD computation, the RGB channel images are divided into 4X4 blocks, after which a further 2X2 blocks division is executed for determination of average pixel intensity. Within the same block, tamper-detection data is entered, whereas the self-recovery data is scattered all over the image utilising inverse and RSA techniques for neighborhood block-based recovery. This method assures easy long-term selfhealing. The outputs produced from testing with 15 multiple host images in varied attacks were constantly performing better with a PSNR ratio upto an average of 45 dB.