AUTHOR=Liu Xiaobao , Xu Biao , Gu Wenjuan , Yin Yanchao , Wang Hongcheng 

TITLE=Plant leaf veins coupling feature representation and measurement method based on DeepLabV3+

JOURNAL=Frontiers in Plant Science

VOLUME=Volume 13 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2022.1043884

DOI=10.3389/fpls.2022.1043884

ISSN=1664-462X

ABSTRACT=The plant leaf veins coupling feature representation and measurement method based on DeepLabV3+ is proposed to solve problems of slow segmentation, partial occlusion of leaf veins, and low measurement accuracy of leaf veins parameters. Firstly, to solve this problem of slow segmentation, the lightweight MobileNetV2 is selected as the extraction network for DeepLabV3+. On the above basis, a Convex Hull-Scan method is applied to repair leaf veins. Subsequently, a Floodfill MorphologyEx Medianblur Morphological Skeleton (F-3MS) refinement algorithm is proposed, reducing the burr phenomenon of leaf veins' skeleton lines. Finally, parameters related to leaf veins are measured. In this study, Mean Intersection over Union (MIoU) and mean Pixel Accuracy (mPA) reach 81.50% and 92.89%, respectively, and the average segmentation speed reaches 9.81 frames per second. Furthermore, the network model parameters are compressed by 89.375%, down to 5.813M. Meanwhile, the measurement accuracy of the leaf vein' length and width reach 96.3642% and 96.1358%, respectively.