Researchers in China have built a deep learning model that can spot the microscopic anatomical signatures of 22 orchid species in nothing more than ground-up stem powder, achieving a mean average precision of 93.35 percent across 65 species-and-structure detection classes. The framework, called YOLO11s-MDOP, was developed by a team at Yunnan University and reported in a study that also introduces a new microscopic image dataset of 6,559 photogra…
This story is only covered by news sources that have yet to be evaluated by the independent media monitoring agencies we use to assess the quality and reliability of news outlets on our platform. Learn more here.