Study on measuring the planting area of winter wheat based on per-field classification of remote sensing

Xiaohe Gu; Yaozhong Pan; Huifang Wang

doi:10.1117/12.833734

30 October 2009 Study on measuring the planting area of winter wheat based on per-field classification of remote sensing

Xiaohe Gu, Yaozhong Pan, Huifang Wang

Author Affiliations +

Proceedings Volume 7498, MIPPR 2009: Remote Sensing and GIS Data Processing and Other Applications; 74981V (2009) https://doi.org/10.1117/12.833734
Event: Sixth International Symposium on Multispectral Image Processing and Pattern Recognition, 2009, Yichang, China

Abstract

According to the problem of spectra variation inside and spectral mixed on boundary of the farmland in mid-resolution images, this paper aims at carrying out per-field classification to improve the accuracy of measuring winter wheat plant area. This paper chooses the urban agriculture region with complex plant structure as experiment area and digitizes the parcel boundary by QuickBird image. By utilizing the farm parcel as end member, the study extracts the information of spectrum, vegetation index and texture from multi-temporal TM images. We establish the evaluation system of field accuracy and total accuracy. The classification methods used in this paper include Support Vector Machine (SVM) and Maximum Likelihood. The study showed that the per-field classification got higher total accuracy field accuracy and stability than per-pixel classification when using for winter wheat plant area measuring. It was useful to improve the accuracy by introducing vegetation index and texture information into per-field classification. The method of both SVM and maximum likelihood got gross accuracy above to 97% and field accuracy above to 90%. The SVM method was more stable than maximum likelihood method, and required much smaller size of training samples. So SVM was more suitable for winter wheat per-field classification. It was useful to improve the accuracy by introducing vegetation index and texture information into per-field classification. This study could provide a new idea about the remote sensing measurement of crop planting area.

Citation Download Citation

Xiaohe Gu, Yaozhong Pan, and Huifang Wang "Study on measuring the planting area of winter wheat based on per-field classification of remote sensing", Proc. SPIE 7498, MIPPR 2009: Remote Sensing and GIS Data Processing and Other Applications, 74981V (30 October 2009); https://doi.org/10.1117/12.833734

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