A method and device for identifying stomata in photos of plant leaves

Abstract

The invention provides a method and device for identifying stomata of plant leaf photos taken under an inverted microscope. The method includes: taking the leaf photo taken under an inverted microscope as an input original image; using the VGG16 convolutional neural network model to convolve the leaf photo Process to obtain the first feature matrix containing the characteristics of the stomata of the blade; map each feature point as the center point back to the original image, and use the point mapped back to the original image as the center point to obtain regions of different sizes; according to the first feature matrix Each feature point and adjacent feature points are subjected to the first preset row and column full convolution processing to determine whether the area mapped back to the original image is the target area of ​​interest and the scope of the target area; map the target area on the original image Go back to the first feature matrix to obtain the corresponding second feature matrix; use the preset fully connected neural network to determine whether the second feature matrix is ​​the stomata area and the range of the stomata area. In this way, the operation of scientific research personnel can be reduced, and the speed and efficiency of scientific research can be improved.

Description

technical field [0001] The invention relates to the field of biotechnology, in particular to a method and device for identifying stomata in photos of plant leaves. Background technique [0002] At present, in the experiments on stomatal research in botany, researchers manually count the number of stomata after taking images, but there are tens of thousands of pictures taken, and it is time-consuming and laborious to count the number of stomata, and the long-term statistics will greatly increase the probability of errors , affecting the progress and efficiency of scientific research. Contents of the invention [0003] The invention provides a method and device for identifying stomata in photos of plant leaves taken under an inverted microscope, which are used to automatically identify the stomatal areas in the photos of plant leaves and make statistics, thereby reducing the operations of scientific researchers and improving the speed and efficiency of scientific research. ...

AI Technical Summary

Problems solved by technology

[0002] At present, in the experiments on stomatal research in botany, researchers manually count the number of stomata after taking images, but there are tens of thousands of pictures taken, and it is time-consuming and laborious to count the number of stomata, and the long-term statistics will greatly increase the probability of errors , affecting the progress and efficiency of scientific research

Images