In this article, we propose a novel dual-filter architecture utilizing RGB-D camera data and dynamic control barrier functions (D-CBFs) for real-Time obstacle avoidance in unstructured environments. The method efficiently handles static, sudden, and dynamic obstacles, maintaining consistent performance across diverse scenarios. To address the challenges of processing substantial pixel and depth data and managing increased optimization solver time in multiobstacle scenarios, we first introduce an enhanced fast-saliency object detection method as the first safety filter for rapid risk area identification. High-risk regions are represented utilizing our enhanced minimal bounding circle (E-MBC) technique, with static and dynamic obstacles differentiated through MBC state comparison and Kalman filtering for obstacle state prediction. This enables efficient online D-CBF construction for each MBC. Subsequently, the second filter establishes buffer zones around established D-CBFs, activating only those corresponding to zones the robot actually enters, rather than all D-CBFs to increase real-Time performance. We prove the system's safety and asymptotic stabilization under this architecture. Simulated and real-world experiments validate our method, demonstrating an RGB-D camera-equipped mobile robot's ability to accomplish tasks while ensuring safety across diverse unknown scenarios.