This paper presents a real-time Automatic Number Plate Recognition (ANPR) system using the ESP32-CAM module combined with cloud-based image processing. The system captures images of vehicle license plates and sends them to a cloud server. Python-based libraries such as OpenCV and EasyOCR process the images to recognize registration numbers. The recognized text is then displayed on an OLED screen connected to the ESP32-CAM. Furthermore, the recognized data is stored in a cloud-based solution, which allows for further integration with vehicle information management systems. Testing results demonstrate strong performance, with character recognition accuracy between 85 and 95% and total processing times of 650 to 950 ms. This includes 50 to 100 ms for image capture, 200–500 ms for network transmission, and 300–350 ms for cloud processing. The system operates effectively at distances of 1 to 3 m, consuming 0.8–1.2 watts. By combining edge and cloud computing, it offers a cost-effective solution for vehicle tracking, traffic management, toll collection and parking monitoring, contributing to Intelligent Transport Systems (ITS) in the IoT domain.