Resources Science ›› 2017, Vol. 39 ›› Issue (8): 1584-1591.doi: 10.18402/resci.2017.08.14

• Orginal Article • Previous Articles     Next Articles

Acreage estimation of mango orchards using object-oriented classification and remote sensing

Chuanshuai REN1,3(), Huichun YE1,2, Bei CUI1,2, Wenjiang HUANG1,2   

  1. 1. Key laboratory of Digital Earth Science,Institute of Remote Sensing and Digital Earth,Chinese Academy of Sciences,Beijing 100094,China
    2. Hainan Key Laboratory of Earth Observation,Sanya 572029,China
    3. University of Chinese Academy of Science,Beijing 100049,China
  • Received:2017-02-10 Revised:2017-07-03 Online:2017-08-20 Published:2017-08-20

Abstract:

China has the second largest mango acreage in the word,and the planting area of mangos has followed a trend of sustained growth. Remote sensing has proved very useful in collecting information about the area of cereals and forest,but has rarely been applied to mango orchards. To utilize remote sensing in mango data collection,we attempted to estimate the acreage of mango orchards in China. Data from SPOT-6 and ASTER GDEM covering the study area was used. The boundary mask and sample points were used for acreage estimation. Vegetation coverage (FVC)and SLOPE factors were used to extract information about mango orchards using the object-oriented method. We found that mango orchard producer’s accuracy obtained by object-oriented classification and FVC and SLOPE data segmentation is as high as 92.81%;user accuracy is as high as 97.19%. Compared with maximum-likelihood classification and object-oriented classifi-cation without FVC and SLOPE data segmentation,object-oriented classification with FVC and SLOPE data segmentation has a higher accuracy for mango orchard extraction:producer accuracy increased by 19.28% and 3.62%,and user accuracy increased by 8.62% and 3.86%,respectively. Therefore,object-oriented classification with auxiliary data improves the accuracy of mango orchard extraction and this method provides an effective technical reference for other artificial orchards.

Key words: mango orchards, object-oriented classification, vegetation coverage, digital elevation model, remote sensing extraction