Leveraging advancements in AI and edge computing, this study develops and implements a low-cost 360-degree blind spot protection system for medium- and large-sized vehicles. The system utilizes entry-level MCUs to perform AI-based image recognition within 11/30 of a second, achieving real-time warning capability without high-end hardware. A self-developed angle detection module enables non-invasive installation without altering vehicle structures. The system integrates wireless cameras, millimeter-wave radar, and ultrasonic sensors to ensure effective detection under low-light or backlit conditions. Multi-modal warning outputs, including sound, light, and vibration, are triggered according to risk levels. Experimental results indicate that the proposed system effectively reduces blind spot risks during turning and reversing, demonstrating both feasibility and cost-effectiveness for vehicle safety applications.