The Transformer model encounters challenges with variable-length input sequences, leading to efficiency and scalability concerns. To overcome this, we propose a pyramid-based hierarchical Spatial-Spectral Transformer (PyFormer). This innovative approach organizes input data hierarchically into pyramid segments, each representing distinct abstraction levels, thereby enhancing processing efficiency. At each level, a dedicated Transformer encoder is applied, effectively capturing both local and global context. Integration of outputs from different levels culminates in the final input representation. In short, the Pyramid excels at capturing spatial features and local patterns, while the Transformer effectively models spatial-spectral correlations and long-range dependencies. Experimental results underscore the superiority of the proposed method over state-of-the-art (SOTA) approaches, achieving overall accuracies of 96.28% for the Pavia University dataset and 97.36% for the University of Houston dataset. Additionally, the incorporation of disjoint samples augments robustness and reliability, thereby highlighting the potential of PyFormer in advancing HSIC. The source code is available at https://github.com/mahmad00/PyFormer .

The Rice crop is considered one of the most widely grown crops in Asia and it is susceptible to various types of illnesses at different stages of production. Food safety and production can be affected by rice plant diseases, as well as a significant decline in the quality and quantity of agricultural goods. Plant diseases can potentially prevent grain harvesting entirely in severe circumstances. As a result, automation of identification and diagnosis of plant disease is widely needed in the agriculture field. Many approaches for doing this problem have been offered with deep learning rising as the preferred method because of its excellent achievement. In this proposed research, we used Hybrid deep CNN transfer learning with rice plant images or the classification and identification of various rice diseases, we employed Transfer Learning to generate our deep learning model using Rice_Leaf_Dataset from a secondary source. The proposed model is 90.8% accurate, Experiments show that the proposed approach is viable, and it can be used to detect plant diseases efficiently and outperformed.

Multihead self-attention and cross-attention mechanisms often suffer from computational inefficiencies, limited scalability, and suboptimal contextual understanding, particularly in hyperspectral image (HSI) classification. These mechanisms struggle to effectively capture long-range dependencies while maintaining computational feasibility due to the quadratic complexity of self-attention. To address these challenges, this work proposes PolicyMamba, a spectral–spatial mamba model enhanced with a localized policy attention mechanism. This mechanism reduces computational overhead by restricting attention to nonoverlapping localized regions and enforcing sparsity constraints, ensuring that only the most informative interactions are retained. A hierarchical aggregation strategy further integrates patch-wise attention outputs, preserving spectral–spatial correlations across scales. In addition, a sliding window patch process enhances local feature continuity while mitigating information loss. The PolicyMamba framework integrates spectral–spatial token generation, token enhancement, localized attention, and state transition modules, significantly improving HSI feature representation. Extensive experiments demonstrate that PolicyMamba achieves superior classification accuracy, outperforming conventional and state-of-the-art methods in land cover classification (LCC) by efficiently modeling intricate dependencies in HSI data.

Hyperspectral image (HSI) classification plays a pivotal role in domains such as environmental monitoring, agriculture, and urban planning. Traditional methods, including conventional machine learning and convolutional neural networks (CNNs), often struggle to effectively capture intricate spectral-spatial features and global contextual information. Transformer-based models, while powerful in capturing long-range dependencies, often demand substantial computational resources, posing challenges in scenarios where labeled datasets are limited, as in HSI applications. To overcome such challenges, this work proposes GraphMamba, a hybrid model that combines spectral-spatial token generation, graph-based token prioritization, and cross-attention mechanisms. The model introduces a novel hybridization of state-space modeling and Gated Recurrent Units (GRU), capturing both linear and nonlinear spatial-spectral dynamics. This approach enhances the ability to model complex spatial-spectral relationships while maintaining scalability and computational efficiency across diverse HSI datasets. Through comprehensive experiments, we demonstrate that GraphMamba outperforms existing state-of-the-art models, offering a scalable and robust solution for complex HSI classification tasks.

On the Internet, every different day, kinds of attacks are deployed on innocent users. Among all, phishing is the most severe attack in which users lose their credentials or private information and their financial status quickly. The attacker uses their credibility or sensitive information to harm the target or victim. The attacker is clever and uses different strategies to fetch user-sensitive information. The existing techniques fail to overcome these issues to some extent. This work focuses on discovering the essential features that help to differentiate the legitimate and illegitimate URLs. We applied a deep learning technique on the benchmark datasets to identify the pattern of phishing URLs. We used gradient boosted decision trees algorithm to train our model and applied the regular deeply connected neural network layers in various sequences and Adam optimizer. The most found patterns will help the system to detect phishing URLs and avoid phishing. We consider the accuracy, Ff-score, and Root Mean Square Error (RMSE) as our evaluation metrics for model evaluation. The results show that the trained model can achieve an approximately 92% accuracy and 94% f-score.