Muhammad Allah Rakha

Muhammad Allah Rakha

Auto-generated profile, based on publicly available information

Muhammad Allah Rakha

Department of Computer Science, FAST-NUCES University, Peshawar, Pakistan

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 January 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
Mahmood AshrafMahmood Ashraf
Mahmood Ashraf
LCLihui Chen
Lihui Chen
XZXichuan Zhou
Xichuan Zhou
MRMuhammad Allah Rakha
Muhammad Allah Rakha
Convolutional neural networks (CNNs) recently have achieved impressive performance for hyperspectral image denoising. However, current CNNs have limitations in exploring spectral correlations across various bands and the interactions among features within each band. Although transformers are introduced to capture spatial-spectral correlation in HSI, generally, they either explore inter-correlation in bands or intra-correlation between bands, neglecting the combination of intra- and inter-correlation in HSI cubes. Besides, transformer methods rarely address hierarchical (i.e., the low- and high-level) features in an adaptive manner. That is, features at different levels are of different importance, whereas these features are tackled equally in these methods. To alleviate these limitations, we introduce a joint architecture (the so-called MAOTformer) of mixed-attention transformers and octave modules for HSI denoising. On the one hand, the mixed attention transformer (MATB) is designed to simultaneously capture pixel relationships inter- and intra-bands by incorporating naive spatial self-attention, bidirectional recurrent channel attention, and progressive channel attention. Besides, a U-net based on mixed attention is equipped with attentive skip connections in MATB, which enables the proposed MAOTformer to explore hierarchical features by U-net and adaptively connect these hierarchical features by the attentive skip connections. On the other hand, we introduce an octave module behind each MATB to utilize multi-scale features for separating noise in high-frequency components. Extensive experiments are conducted on synthetic and real-world hyperspectral images (HSI), showing that the proposed method outperforms state-of-the-art methods.