A Framework to Automatic Detect Center Pivots Using Land Use and Land Cover Data

Marcos Lima Rodrigues,Thales Sehn Körting,Gilberto Ribeiro De Queiroz

doi:10.14393/rbcv73n4-60553

Marcos Lima Rodrigues, Thales Sehn Körting + Show 1 more

Open Access

https://doi.org/10.14393/rbcv73n4-60553

Copy DOI

Abstract

Water management is a key field to support life and economic activity nowadays. The greatly increased mechanization of agriculture, mainly through center pivot irrigation systems, represents a big challenge to control this resource. Irrigated agriculture makes up the large majority of consumptive water use, therefore it is important to identify and quantify these systems. Currently, with 6.95x10⁶ ha, Brazil is among the 10 largest countries in irrigation areas in the world. In this study, a combined Computer Vision and Machine Learning approach is proposed for the identification of center pivots in remote sensing images. The methodology is based on Circular Hough Transform (CHT) and Balanced Random Forest (BRF) classifier using vegetation indices NDVI/SAVI generated from Landsat 8 images and Land Use and Land Cover (LULC) data provided by project MapBiomas. The candidate's circles of pivots identified on images are filtered based on vegetation behavior and shape characteristics of these areas. Our approach was able to detect 7358 pivots, reaching 83.86% of Recall for 52 scenes analyzed overall Brazil compared with mapping done by the Brazilian National Water and Sanitation Agency (ANA). In some scenes, the Recall reaches up to 100%.

Highlights

The practice of irrigation is one of the oldest techniques in agricultural production, used mainly by civilizations that developed in arid regions such as Egypt and Mesopotamia (BRITANNICA ESCOLA WEB, 2020)
The proposed methodology enables automatic identification and quantification of center pivots systems over large regions in time and space based on target detection on remote sensing images of medium spatial resolution using digital processing techniques and machine learning (ML) models
The multistage method uses the temporal composition of vegetation indices to obtain soil adjusted vegetation index (SAVI) amplitudes and masks of crop areas from land use and land cover (LULC) data to highlight areas with a dynamic crop activity, thereby reducing the effort and cost of not classifying targets from areas potentially not related to irrigated areas

Summary

Introduction

The practice of irrigation is one of the oldest techniques in agricultural production, used mainly by civilizations that developed in arid regions such as Egypt and Mesopotamia (BRITANNICA ESCOLA WEB, 2020). From 1970 to 1980, there was a significant intensification of agricultural activity and mechanization in other regions, driven mainly by government programs such as the Northeast Irrigation Program (PROINE), Irrigation Equipment Financing Program (PROFIR) and National Irrigation Program (PRONI) (GUIMARÃES; LANDAU, 2014). This contributed to new national irrigation poles, which are special areas for water resource management for irrigated agriculture on a national scale because for the total irrigated area, concentration and growth are observed over the short and medium time (ANA, 2019). This type of system irrigates a circular area by rotating this structure around a fixed point, called a pivot point, which serves to anchor the system and extract the water (MARANHA, 2018)

Objectives

Methods

Results

Discussion

Conclusion