Loading…
US-IALE 2018 has ended
Back To Schedule
Wednesday, April 11 • 4:15pm - 4:30pm
REMOTE SENSING/IMAGE ANALYSIS: Improving Crop Identification and Acreage Estimation Using Crop Landscape Models

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!

AUTHORS: Qiangzi Li*, Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences

ABSTRACT: Rapid, accurate, and reliable crop acreage estimation is significant for the macro cropping structure adjustment and the national agricultural policy development. Remote sensing techniques can provide substantive and timely crop information, and have been used in crop identification and crop acreage estimation since 1970s. It has been found that crop classification accuracies rely greatly on the crop distribution patterns in the study area. This study developed crop landscape models to quantify crop distribution and use these models to improve the accuracy of crop identification and acreage estimation. The effects of crop spatial distribution on crop identification and acreage estimation with remote sensing were explored in two experimental sites in eastern Ontario, Canada and northeastern China. Firstly, exploratory analysis was conducted to investigate the influence of classification features and classifiers on the classification accuracy, and the optimal features and classifiers were identified. Secondly, four landscape metrics were selected based on conceptual crop landscape models, and six landscape sub-regions were delineated accordingly. Lastly, optimal classification features and classifiers were determined for each landscape sub-region, in order to develop improved crop acreage estimation models. The improved crop acreage estimation models were applied in Hunan province, China for regional rice acreage estimation. The results demonstrated that acreage accuracy can be increased in five out of six landscape sub-regions, by approximately 4% in sub-region 3, 5, and 6, and by over 6% in sub-region 1 and 2.

Wednesday April 11, 2018 4:15pm - 4:30pm CDT
Water Tower Parlor