This paper presents an automated classification system for Areca nut, utilizing image processing to assess features such as texture, colour, and size. The system is built on a hardware configuration that includes an ESP32 microcontroller, a camera, and a motorized sorting mechanism. The areca nut images are captured and sent to a cloud-based server via the ESP, where a machine learning algorithm processes the data to classify the areca nut based on predefined categories. Once the classification is complete, the results are relayed back to the ESP, which then activates motors to sort each areca nut item into the appropriate bin. This approach is both efficient and scalable, providing a robust solution for real-time, automated sorting, crucial in agriculture processing industries. By integrating edge-cloud technologies, the system achieves high classification accuracy while remaining cost-effective and adaptable to various environmental conditions. The research highlights the potential of IoT and machine learning applications in agricultural automation, paving the way for enhanced productivity and reduced manual labour in areca nut sorting. Experimental results demonstrate the system's effectiveness, showing accurate classification rates and efficient sorting in real-world testing scenarios. This solution could be adapted for other agricultural products, offering a generalizable framework for IoT-based, machine-learning-driven classification and sorting systems.