Robust Image Recognition using Quantum-Inspired Algorithms on Low-Power Electronic Platforms

Abstract

Image recognition has become an essential component of artificial intelligence (AI) applications, from autonomous vehicles to smart surveillance systems. However, conventional deep learning models such as convolutional neural networks (CNNs) require significant computational resources, making them inefficient for low-power or embedded devices. This research introduces a quantum-inspired computing framework for image recognition designed to function effectively on low-power electronic platforms. By leveraging quantum annealing principles, amplitude encoding, and hybrid optimization techniques, the study proposes models that achieve high recognition accuracy with minimal energy consumption. Experimental simulations conducted on embedded GPUs and neuromorphic chips demonstrate that quantum-inspired algorithms outperform classical CNNs in energy efficiency by up to 45% while maintaining competitive accuracy levels above 92% across multiple datasets. This paper highlights how quantum-inspired methodologies can bridge the gap between computational efficiency and accuracy in next-generation AI systems.