One of the key factors in maximizing agricultural productivity with minimum cost and environmental impact is efficient use of resource management. Accurate detection of plant maturity plays a crucial role in this management, particularly for tasks related to harvesting, maturity classification and crop health assessment. This study explores the application of YOLO (You Only Look Once) based detection models for smart greenhouses, with a specific focus on evaluating the performance of five YOLOv8 models: nano (YOLOv8n), small (YOLOv8s), medium (YOLOv8m), large (YOLOv8l), and extra-large (YOLOv8x) - to classify tomato maturity stages into ripe and unripe, using a set of images. The results showed a good improvement in accuracy for all models. The YOLOv8n and YOLOv8s achieved 0.827% for mean Average Precision (mAP) with an Intersection of Union (IoU) of 0.5. While YOLOv8l attains 0.821%. The YOLOv8x reached 0.815% and YOLOv8m accomplished 0.838%. From the results, YOLOv8m displayed strong potential for real-time fruit ripeness classification.