Research on farmland boundary extraction based on deep learning and edge detection operator for GF-2 data

Yan Wang; Lingjia Gu; Haixia Yan; Mingda Jiang

doi:10.1117/12.2593040

1 August 2021 Research on farmland boundary extraction based on deep learning and edge detection operator for GF-2 data

Yan Wang, Lingjia Gu, Haixia Yan, Mingda Jiang

Proceedings Volume 11829, Earth Observing Systems XXVI; 1182917 (2021) https://doi.org/10.1117/12.2593040
Event: SPIE Optical Engineering + Applications, 2021, San Diego, California, United States

Conference Poster

Abstract

With the application of agriculture in remote sensing technology, it has become a development trend to extract accurate farmland boundary information from high-resolution remote sensing images. Farmland plot is the smallest unit of crop planting, so accurate boundary extraction of farmland plot is of great significance for crop classification. Jilin Province in Northeast of China is selected as the experimental area. Taking GF-2 data as the data source, the effects of UNet, HED algorithm based on deep learning, CNN color edge extraction algorithm and traditional morphological edge detection operator in farmland boundary extraction are compared, and then a farmland boundary extraction method based on the combination of deep learning and morphological edge detection operator is proposed in this paper. Compared with using deep learning network or morphological edge detection operators alone, this method has higher peak signal-to-noise ratio and structural similarity, and better farmland edge extraction effect.

Citation Download Citation

Yan Wang, Lingjia Gu, Haixia Yan, and Mingda Jiang "Research on farmland boundary extraction based on deep learning and edge detection operator for GF-2 data", Proc. SPIE 11829, Earth Observing Systems XXVI, 1182917 (1 August 2021); https://doi.org/10.1117/12.2593040

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