https://archive.conscientiabeam.com/index.php/76 Conscientia Beam en-US Review of Computer Engineering Research 2412-4281 https://archive.conscientiabeam.com/index.php/76/article/view/3706 <p>This research examines a multi-attention residual integrated network with an enhanced fireworks algorithm for remote sensing image classification. Remote sensing (RS) picture classification is important for land cover mapping, environmental monitoring, and urban planning. Remote sensing image classification is important in earth observation since the military and commercial sectors have focused on it. Due to RS data's high complexity and limited labelled examples, classifying RS pictures is difficult. Deep Learning (DL) techniques have made great strides in RS image categorization, expanding this field's potential. This research introduces Multi-Attention Residual Integrated Network with Enhanced Fireworks Algorithm (MAR-EFA) to improve hyper spectral image identification. MARIN-EFA improves feature fusion and removes unneeded features to overcome technique constraints. The suggested method weights features using different attention models. These characteristics are then carefully extracted and integrated using a residual network. Final contextual semantic integration on deeply fused features is done with a Bi-LSTM network. Our population-based Enhanced Fireworks Algorithm (EFA) is inspired by fireworks' explosive performance and optimises MARIN parameters. Attention techniques and an improved optimisation algorithm improve performance over current systems. Numerous Eurosat dataset studies were assessed using various performance indicators. The simulation results show that MARIN-EFA outperforms current methods. The suggested technique shows promise for improving RS picture classification and allowing more accurate and reliable data categorization.</p> Josephine Anitha Antony Gladis Dennis Copyright (c) 2024 2024-04-05 2024-04-05 11 2 58 72 10.18488/76.v11i2.3706 https://archive.conscientiabeam.com/index.php/76/article/view/3778 <p>This research aims to enhance the efficiency of simulating 3D cameras utilizing Photonic Mixer Devices (PMD) technology, crucial for applications in computer vision, robotics, and augmented reality. Despite their significance, the computational demands of simulating PMD 3D cameras present substantial challenges in time and resource management. This study proposes a novel approach to optimizing simulation time without sacrificing accuracy, achieved through advanced algorithms and parallel computing techniques. Through a comprehensive analysis of existing simulation methodologies, bottlenecks are identified, and tailored optimization techniques are implemented. The system is designed to simulate PMD sensors, wherein ray tracing precedes power calculation, essential for determining pixel radiance and irradiance. However, the inherent computational intensity of the sequential power calculation algorithm presents a challenge of speed, particularly for PMD sensor simulation reliant on fast-imaging technology. To address this issue, a parallel algorithm leveraging General Purpose Graphics Processing Units (GP GPUs) is proposed and implemented. Experimentation is carried out on Volta (GV100) Graphics Processing Unit (GPU) with varying block sizes from 32 to 1024 in the multiples of 32. Experimental results demonstrate significant speed enhancements, with a maximum speed up of 78% utilizing Volta GPU with a block size of 1024, thereby showcasing the efficacy of the proposed methodology.</p> Sangita Gautam Lade Sanjesh Pawale Aniket Patil Copyright (c) 2024 2024-06-20 2024-06-20 11 2 73 84 10.18488/76.v11i2.3778