<![CDATA[SBL (Single Bottom Labeller) machine which was still not precise for placing the bottom label batches of lip cream products. Postal operators must then check one by one to ensure product quality control so that there are bottlenecks. Therefore, this research uses the YOLOv5 deep learning method to solve this problem. The YOLOv5 algorithm uses the idea of regression, making it easier to learn generalizations, target characteristics and solve speed problems. The YOLO algorithm uses a one-stage neural network to complete the localization and classification of detected objects in real time. The core idea of YOLO is to take the entire image as network input and directly return the bounding box position and bounding box class in the output. In YOLO, each bounding box is predicted by the features of the entire image, and each bounding box contains the five predictions and a confidence, which is relative to the grid cell in the middle of the bounding box. The results of research using the YOLOv5 method can classify into three classification categories of bottom label placement, namely accept, reject and no label. In addition to classifying, this study will also send triggers to the controller if the bottom label is said to be rejected and no label to do product sorting to assist QC operators in sorting lip cream production defects.]]>
Copyrights © 2023
