End-to-end directional peanut stomatal phenotype real-time intelligent detection method and system

The invention discloses a real-time intelligent detection method and a real-time intelligent detection system, namely Peanut Net, for the phenotype of leaf stomata of an end-to-end directional peanut L. The core of the invention is an end-to-end directional peanut leaf stomata phenotype real-time intelligent detection method. According to the method, aiming at peanut leaf epidermis stomata image data obtained in a lossy or lossless mode, real-time peanut leaf stomata positioning and full-automatic obtaining of phenotype information such as density and opening degree are achieved in a one-stop mode by constructing an end-to-end directional target detection network. The PeanutNet mainly comprises the following modules: a peanut leaf image import module, a peanut leaf stomata intelligent detection module, a batch processing and display module, a model parameter adjustment module, a detection and identification detailed information storage module and the like. The system has real-time performance, great convenien