The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLII-3/W12-2020
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W12-2020, 73–78, 2020
https://doi.org/10.5194/isprs-archives-XLII-3-W12-2020-73-2020
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W12-2020, 73–78, 2020
https://doi.org/10.5194/isprs-archives-XLII-3-W12-2020-73-2020

  04 Nov 2020

04 Nov 2020

DETECTING CENTER PIVOTS IN MATOPIBA USING HOUGH TRANSFORM AND WEB TIME SERIES SERVICE

M. L. Rodrigues1, T. S. Körting2, G. R. de Queiroz2, C. P. Sales1, and L. A. R. da Silva1 M. L. Rodrigues et al.
  • 1Graduate program in Applied computing, National Institute for Space Research, São José dos Campos, SP, Brazil
  • 2Image Processing Division (DPI), National Institute for Space Research, São José dos Campos, SP, Brazil

Keywords: Center pivots, Hough Transform, Landsat, MODIS, Time series

Abstract. In the last decades, the Brazilian Cerrado biome has undergone major transformations due to the expansion of the agricultural frontier. The region called MATOPIBA acronym for states Maranhão, Tocantins, Piauí, and Bahia can be considered very attractive for agricultural expansion. The Cerrado predominates in the MATOPIBA region (91% of the area), also having small areas of the Amazon and Caatinga biomes to the northeast and east, respectively. In this work, we will present a study to identify center pivot irrigation systems in the MATOPIBA region using remote sensing images from Landsat-8 satellite. The methodology is based on the use of robust edge detection techniques such as Canny, Circular Hough Transform (CHT) and time series extraction through the Moderate Resolution Imaging Spectroradiometer (MODIS) product MOD13Q1 which has two vegetation indices NDVI and EVI. These time series will be used to filter the detected circles, seeking to eliminate the circles that do not correspond to center pivots. Our approach detected 80% of the center pivots mapped by the Brazilian National Water Agency (ANA) used as a knowledge base. The states with better detection were Piauí and Bahia that showed the accuracy of 90% and 85% respectively, Maranhão obtained 57% and Tocantins 41%.