The invention relates to the technical field of
computer vision and
deep learning, and provides a
microscopic image-based gramineous
plant leaf stomatal index measurement method, which comprises the following steps: 1, acquiring a
microscopic image of a gramineous
plant leaf to be measured; 2, constructing and training an air hole recognition model based on a
deep learning target detection
algorithm; 3, constructing and training a
cell network prediction model based on a
deep learning semantic segmentation
algorithm; 4, obtaining the total number of air holes by utilizing the air hole recognition model, utilizing the
cell network prediction model to obtain a
cell network prediction image, sequentially carrying out self-adaptive threshold binarization
processing, skeleton extraction, morphological operation of firstly corroding and then expanding, connected domain counting and connected domain filtering on the
cell network prediction image to obtain the number of cells, and calculatingstomatal indices. According to the invention, the precision and efficiency of stomatal index measurement can be improved, and the technical problem of misjudgment of stomata and cells due to the timeand labor waste and subjectivity in the prior art is effectively solved.