Accurate and efficient human counting is essential for optimizing public transportation and advancing smart city infrastructure. This paper evaluates proposed lightweight deep learning models for autonomous human counting system on low-cost hardware, ensuring real-time monitoring and enhanced operational efficiency. While existing methods, such as DeepSORT, Kalman Filters, and YOLO variants, are often implemented on high-end hardware, they typically prioritize accuracy over computational efficiency. Few object detection and tracking techniques can run in realtime on low-end hardware. This work advances the field by utilizing optimized deep learning models suitable for embedded systems with constrained resources. Specifically, fine-tuned YOLOv8 is employed for head detection, combined with ByteTrack for robust tracking, outperforming YOLOv5 and YOLOv11 in accuracy and efficiency. Archiving the 15 FPS and more then 90％ accuracy on the real environment deployment on both RISC-V architecture with an integrated NPU (MaixCAM) and ARM v8 (Raspberry Pi), The proposed system demonstrates its suitability for real-time, cost-effective, and scalable autonomous human counting in public transit environments.