A clump tree instance segmentation based on PhenoCam imagery using MaskRCNN method

Mengying Cao; Qinchuan Xin

doi:10.1117/12.2667381

10 February 2023 A clump tree instance segmentation based on PhenoCam imagery using MaskRCNN method

Mengying Cao, Qinchuan Xin

Proceedings Volume 12552, International Conference on Geographic Information and Remote Sensing Technology (GIRST 2022); 1255220 (2023) https://doi.org/10.1117/12.2667381
Event: International Conference on Geographic Information and Remote Sensing Technology (GIRST 2022), 2022, Kunming, China

Abstract

Vegetation is one of the important land covers and plays an important role in the ecosystem and climate change. Most of the current studies on vegetation and atmospheric change are based on the material and energy exchange between vegetation canopy and atmosphere. The study of the interaction between individual vegetation and atmosphere is affected by the limitations of acquisition. The development of the convolutional neural network derives a new approach for obtaining clump trees from PhenoCam. In this paper, a MaskRCNN method is proposed to train the model from PhenoCam images. Image enhancement processing is added to improve the training accuracy of the model. Then the training model is used to instance segment a clump tree from an individual PhoneCam imagery. The result of MaskRCNN instance segmentation mAP (mean average precision) can reach 0.887. In the vegetation boundary area and complex vegetation types, the effect of MaskRCNN segmentation is better than other methods. With the increase of image enhancement processing, the training accuracy of the model and the robustness of the data are enhanced. This study provides a feasible and efficient acquisition method for the clump tree segmentation.

Citation Download Citation

Mengying Cao and Qinchuan Xin "A clump tree instance segmentation based on PhenoCam imagery using MaskRCNN method", Proc. SPIE 12552, International Conference on Geographic Information and Remote Sensing Technology (GIRST 2022), 1255220 (10 February 2023); https://doi.org/10.1117/12.2667381

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
6 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Image segmentation

Vegetation

Image processing

Education and training

Data modeling

Image enhancement

Point clouds

Show All Keywords

Keywords/Phrases

Search In:

Publication Years