The determination of the best breed among the current samples was heavily dependent on its phenotyping traits. An innovative autonomous screening approach to enhance efficiency and accuracy in phenotyping of plant breeding is a recent trend. In this study, tomato seedlings of the “Chena” variety, known for their resistance to bacterial wilt, were planted and germinated under controlled conditions. Images of seedlings, captured and were processed using YOLOv8, a high-accuracy object detection model, to extract features such as leaf area, number of leaves, and seedling height. Data augmentation techniques enhanced the dataset for training machine learning models, including K-Nearest Neighbors (KNN), Support Vector Machines (SVMs), Decision Trees, and Deep Neural Networks (DNNs). The best-performing model was used to predict vegetative growth, facilitating the identification of the most productive and resilient seedlings. The best result was achieved by KNN, with an accuracy of 83%. Overall, the results highlight its superior performance compared to traditional methods in terms of speed, accuracy, and scalability. Ultimately, the methodology recommended in this research is of paramount importance for advancing plant breeding practices in future research endeavors.