A Novel CNN Approach for Efficient Areca Nut Sorting Machine

Manual sorting of areca nuts based on quality not only consumes valuable time but also leads to financial losses for farmers due to labor wages. Proposed work explores the implementation of neural networks and image processing techniques for accurate detection and classification of areca nut quality. To ensure an effective sorting process based on their quality attributes, we utilize a back-propagation neural network classifier. We achieved an impressive classification accuracy of 94.9%. This outcome demonstrates the robustness and precision of the system in accurately differentiating between various grades of areca nuts based on their quality attributes. Further, we developed an automated areca nut Sorting Machine using Raspberry Pi for efficient and precise sorting of areca nuts based on our previously determined classifications. The machine is structured with two distinct phases of separation, primarily focusing on size and color differentiations. The machine demonstrates its capability to bear a load of up to 10 kg while achieving a bagging efficiency of 700 areca nuts per revolution of the belt. These enhanced design features ensure the machine's durability, efficiency, and overall performance, making it well-suited for its intended application.