Deep Learning-Based Approach for Identifying and Counting Wooden Blocks with YOLO

The wood processing industry in Indonesia, especially in the Southwest Papua region, faces ongoing challenges in accurately counting wooden logs, a task traditionally performed manually. Manual methods are time-intensive and prone to error, leading to inefficiencies in operations and weaknesses in resource management. This study addresses these challenges by applying a deep learning-based object detection approach, specifically the You Only Look Once version 8 (YOLOv8) algorithm, to automate the detection and counting of wooden beams in real time. YOLOv8 was chosen for its ability to perform high-speed and accurate detection even under varying environmental conditions, such as different lighting levels and camera angles. The model was trained on a custom dataset consisting of 265 annotated images of wooden beams, with a split of 70% for training, 20% for validation, and 10% for testing. Performance evaluation using a confusion matrix revealed a detection accuracy of 94%. These findings suggest that YOLOv8 is highly effective in supporting automation within wood processing workflows. By reducing dependency on manual labor and minimizing counting errors, the system contributes to more accurate inventory tracking and enhanced productivity. This research demonstrates the potential of integrating AI-driven models into mobile and industrial applications for improved efficiency in forestry-related sectors.