International Journal of Wireless and Ad Hoc Communication IJWAC 2692-4056 10.54216/IJWAC https://www.americaspg.com/journals/show/3740 2019 2019 Assistant Professor, Department of ECE, Knowledge Institute of Technology, Salem, Tamil Nadu, India Chandraman Chandraman Assistant Professor, Department of ECE, Knowledge Institute of Technology, Salem, Tamil Nadu, India Chandraman. M. Professor, Department of ECE, Knowledge Institute of Technology, Salem, Tamil Nadu, India Santhiyakumari. N. Associate Professor, Department of ECE, Knowledge Institute of Technology, Salem, Tamil Nadu, India Saravanan. V. Associate Professor, Department of ECE, Knowledge Institute of Technology, Salem, Tamil Nadu, India Shanmugasundaram. P. PG Scholar, Department of ECE, Knowledge Institute of Technology, Salem, Tamil Nadu, India Arun. A. Artificial intelligence (AI) is becoming more and more necessary for devices, particularly for network edge image processing applications. Building Very-Large-Scale Integration (VLSI) systems that are specifically tuned for low power consumption and enable edge AI techniques for real-time image processing is the aim of this research. One of Edge AI's key characteristics is its ability to process data and make judgements instantly. Edge AI reduces latency by eliminating the need to move massive amounts of data from one location to the cloud. Quick response times are made feasible, which is essential for applications such as industrial automation and autonomous driving. The study will investigate hardware accelerators and approximation computing as efficient approaches to perform image processing algorithms on low-resource edge devices. If all created data were transferred to the cloud, the network infrastructures would be overwhelmed by the exponential growth in linked devices. Edge AI solves this issue by significantly reducing the amount of data that needs to be sent across the network by doing computations locally. This increases the scalability of AI systems and decreases operating costs associated with data transport. By using custom VLSI design, the project aims to achieve significant energy savings over traditional software-based solutions. This will pave the way for edge AI to be widely applied in battery-powered devices for longer battery life and tasks like object and picture identification. 2025 2025 01 11 10.54216/IJWAC.090201 https://www.americaspg.com/articleinfo/20/show/3740