http://bura.brunel.ac.uk/handle/2438/13037 2026-04-21T05:33:51Z 2026-04-21T05:33:51Z Alhumaima, RS Al-Karawi, Y Al-Raweshidy, H http://bura.brunel.ac.uk/handle/2438/33048 2026-04-02T13:24:01Z 2026-01-01T00:00:00Z

Title: Cybersecure Synchronization of Entangled Quantum Neural Networks with RIS and QKD for 6G Holographic Communications Authors: Alhumaima, RS; Al-Karawi, Y; Al-Raweshidy, H Abstract: ... Description: ...

2026-01-01T00:00:00Z Min, C Lei, T Wang, X Wang, Y Meng, H Nandi, AK http://bura.brunel.ac.uk/handle/2438/33028 2026-03-24T03:00:38Z 2026-03-12T00:00:00Z

Title: Mitigating model coupling in semi-supervised segmentation via deep non-consistent mean teacher and fully collaborative learning Authors: Min, C; Lei, T; Wang, X; Wang, Y; Meng, H; Nandi, AK Abstract: Semi-supervised learning methods based on teacher-student frameworks have achieved remarkable success in image segmentation. However, popular teacher-student models are prone to early subnet coupling, which limits segmentation performance. Moreover, most existing approaches rely on strong-weak perturbation schemes for consistency learning, overlooking peer-level supervision between different perturbations and failing to fully exploit the potential information from unlabeled data. To address the above issues, we propose ComMatch, a novel semi-supervised image segmentation method built upon deep non-consistency and fully collaborative learning. Specifically, a deeply non-consistent mean-teacher structure is designed, which expands the learning space by constructing deep inconsistencies at both the data and network levels within a multi-stream learning framework and can effectively alleviate the problem of early subnet coupling. Meanwhile, to maximize the latent information from unlabeled data, a fully collaborative learning strategy is proposed, which explores the necessity of peer-level loss under deep inconsistency perturbations and further combines cross-level and peer-level losses to deeply mine the latent knowledge from unlabeled data. Experimental results show that the proposed ComMatch method surpasses the current state-of-the-art methods, achieving segmentation accuracies of 78.68% and 77.89% (1/16) respectively in the mIoU metric on the PASCAL VOC and Cityscapes datasets. Code is available at https://github.com/Minchongdan/ComMatch. Description: Data availability: Data will be made available on request.

2026-03-12T00:00:00Z Alturfi, AM Salgotra, R Almajidi, SD Hussein, RA Al-Raweshidy, H http://bura.brunel.ac.uk/handle/2438/32908 2026-03-02T03:00:42Z 2026-01-01T00:00:00Z

Title: A Novel Modified Sine Cosine Algorithm for Reducing Side lobe Level of Linear Antenna Array Authors: Alturfi, AM; Salgotra, R; Almajidi, SD; Hussein, RA; Al-Raweshidy, H Abstract: ... Description: ...

2026-01-01T00:00:00Z Song, H Han, J Ma, H Jia, H Shen, X Gou, J Lai, Y Meng, H http://bura.brunel.ac.uk/handle/2438/32740 2026-04-14T08:09:06Z 2025-12-31T00:00:00Z

Title: Edge priors guided deep unrolling network for single image super-resolution Authors: Song, H; Han, J; Ma, H; Jia, H; Shen, X; Gou, J; Lai, Y; Meng, H Abstract: The field of single image super-resolution (SISR) has garnered significant consideration over the past few decades. The primary challenge of SISR lies in restoring the high-frequency details in low-resolution (LR) images, which are crucial for human perception. Current deep learning-based SISR methods either increase the depth of the network or indiscriminately incorporate emerging technologies, such as attention mechanisms, to address this challenge. However, these methods treat the deep networks as a black-box, achieving performance at the cost of efficiency and network redundancy, without carefully considering how internal components interact to enhance the reconstruction quality. To address this limitation, we incorporate edge priors into a classical image restoration model to design network framework and propose an edge priors guided deep unrolling network (EPGDUN), which consists of three components: Edge Feature Extraction Module (EFEM), Intermediate Variable Updating Module (IVUM), and Variable-Guided Reconstruction Module (VGRM). Specifically, we unroll the image restoration model with edge priors via a half-quadratic splitting and proximal gradient descent method to gain three subproblems, whose solving process corresponds to the iterative stages of the three submodules of EPGDUN. The combination of edge priors constrains the network output, strengthening the image boundaries, enhancing interpretability, and improving the network’s understanding of image structure. Extensive experiments illustrate that EPGDUN achieves performance on par with or exceeding that of the state-of-the-arts, uninterpretable black-box neural models and interpretable deep unrolling networks. These findings underscore EPGDUN’s potential to advance low-level vision applications and other domains requiring mathematical interpretability. The source code for EPGDUN will be available at https://github.com/songhp/EPGDUN. Description: Highlights: • We propose an interpretable EPGDUN for single image super-resolution. • We design a non-local block to gather a broader range of prior information. • We develop a cross-fusion module to selectively fuse edge and image features. • Our model demonstrates outstanding qualitative and quantitative performance.; Data availability: Data will be made available on request.

2025-12-31T00:00:00Z