Identification of winter canola based on dual-polarized SAR datasets in hilly mountainous areas of southwest China.

SENTINEL-1 (Artificial satellite); CANOLA; SYNTHETIC aperture radar; SUPPORT vector machines; FORESTS & forestry; IDENTIFICATION; CROPS

To improve the accuracy of canola identification in the southwest cloudy and foggy mountains of China, this letter puts forward a group of canola recognition features with the multi-temporal Sentinel-1 images. The Jeffries-Matusita (J-M) distance of sample types and the accuracy of canola extraction of support vector machine were compared. The results showed that the J-M values of canola crop with forest land and other green vegetation were improved obviously, among which the J-M value of forest land category increased to 1.93, and that of other green vegetation 1.82. The producer's accuracy (PA) of canola was improved to 70.93%, and the user's accuracy (UA) was increased to 75.86% with F1-measure 73.31%. The canola crop area recognition accuracy was 82.58%. The feature combination can enhance the distinguishability among sample categories, improve the extraction accuracy of canola crops, but the results still have a gap with the reliable results proposed by previous scholars due to the complex crops planting structure, the irregularity of the plot and the resolution of Synthetic Aperture Radar (SAR) images. This research can provide a reference for the distribution extraction of canola in mountain areas. [ABSTRACT FROM AUTHOR]

Copyright of Remote Sensing Letters is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)