Exécution de notre outil de synthèse sur un nouvel article
Vérifiez l’article ci-dessous, puis lancez sa synthèse.
AI-Powered Seed Viability Assessment in Hydroponic Systems: A Comparative Review of Computer Vision and Deep Learning Approaches
Hemlata Parmar (Manipal Academy of Higher Education), Parth Gupta (Manipal Academy of Higher Education), Singh Manas Mukundkumar (Manipal Academy of Higher Education), Abhinav Mukherjee (Manipal Academy of Higher Education)
Résumé : This review paper provides a comparative review of research studies that focused on AI computer vision models for the detection of ungerminated seeds in hydroponic systems. Hydroponic farming is becoming increasingly popular as operating a more sustainable form of agriculture unlike traditional farming practices; nevertheless, challenges remain for efficiency in resource use. The identification and early detection of non-viable saplings allows growers to utilize their water, nutrients, and energy for viable products while improving sustainability overall. This paper details the methods that were employed in studies found in recent literature which focused on CNNs, served with hyperspectral imaging, hybrid DL models, along with a comparison of performance metrics, dataset qualities and practical application. Furthermore, potential future research directions directed at continuing to enhance AI-based seed viability assessment and provide options for real-time decision making in this area are delivered.