Study of land use classification in an arid region using multispectral satellite images

Abstract. This study explores the short-, medium- and long-term impacts of expansion of the sugarcane plantation on the water balance of the Rio Grande Basin, Brazil, as estimated by changes in evapotranspiration, soil moisture content and surface runoff calculated by a hydrological model. Twenty years of simulation are made using three different land use scenarios that include the basin area planted with sugarcane in 1993, 2000 and 2007 as estimated from satellite images. Complementary, it is used a scenario for sugarcane plantation defined by the Brazilian Institute for Agricultural Research (EMPRAPA) as all areas suitable for sugarcane cultivation within the Rio Grande Basin. In addition, parameters for sugarcane fields were specifically defined via calibration and validation of the hydrological model for all growth phases based on the annual cycle of sugarcane phenology in the Rio Grande Basin. According to results from the land use classification of satellite images, the expansion of sugarcane fields mostly replaced pasture lands. Modelling results for short-, medium- and long-term clarify that impacts of this expansion depended not only on the amount of areas planted with sugarcane, but also the type of land use replaced, location of the expansion within the basin and regional soil properties. Largest impacts on the water balance are observed if areas located close to headwaters with low soil water capacity are planted with sugarcane. In case all areas suitable for sugarcane plantation, as defined by EMBRAPA will actually be planted, simulations showed that the annual accumulated values of evapotranspiration increase up to 180% while surface runoff is reduced to 20% of the values calculated using a land use scenario from 1993.

Download Full-text

Land Classification using Convolutional Neural Networks

International Journal of Recent Technology and Engineering - 2 ◽

10.35940/ijrte.a3030.079220 ◽

2020 ◽

Vol 9 (2) ◽

pp. 79-83

Keyword(s):

Neural Networks ◽

Land Use ◽

Convolutional Neural Networks ◽

Satellite Images ◽

Residential Buildings ◽

Classification Performance ◽

Optimization Techniques ◽

Physical Aspect ◽

Land Use Classification ◽

Forest Lands

Identifying the physical aspect of the earth’s surface (Land cover) and also how we exploit the land (Land use) is a challenging problem in environment monitoring and much of other subdomains. One of the most efficient ways to do this is through Remote Sensing (analyzing satellite images). For such classification using satellite images, there exist many algorithms and methods, but they have several problems associated with them, such as improper feature extraction, poor efficiency, etc. Problems associated with established land-use classification methods can be solved by using various optimization techniques with the Convolutional neural networks(CNN). The structure of the Convolutional neural network model is modified to improve the classification performance, and the overfitting phenomenon that may occur during training is avoided by optimizing the training algorithm. This work mainly focuses on classifying land types such as forest lands, bare lands, residential buildings, Rivers, Highways, cultivated lands, etc. The outcome of this work can be further processed for monitoring in various domains.

Download Full-text

Integrated Landscape Pattern and Classification Tree for Land Use Classification in Inland River Basin of China Arid Region

IGARSS 2008 - 2008 IEEE International Geoscience and Remote Sensing Symposium ◽

10.1109/igarss.2008.4779132 ◽

2008 ◽

Author(s):

Yuan Qi ◽

Jianhua Wang ◽

Zhen Xu ◽

Changwei Lin

Keyword(s):

Land Use ◽

River Basin ◽

Landscape Pattern ◽

Arid Region ◽

Classification Tree ◽

Land Use Classification ◽

Inland River ◽

Inland River Basin

Download Full-text

Comparison of Accuracy Measures for RS Image Classification using SVM and ANN Classifiers

International Journal of Electrical and Computer Engineering (IJECE) ◽

10.11591/ijece.v7i3.pp1180-1187 ◽

2017 ◽

Vol 7 (3) ◽

pp. 1180

Author(s):

S.V.S. Prasad ◽

T. Satya Savithri ◽

Iyyanki V. Murali Krishna

Keyword(s):

Land Use ◽

Land Use Planning ◽

Satellite Images ◽

Support Vector ◽

Svm Classifier ◽

Landsat 8 ◽

Hybrid Technique ◽

Multispectral Satellite Images ◽

Artificial Neural Network Ann

<p>The accurate land use land cover (LULC) classifications from satellite imagery are prominent for land use planning, climatic change detection and eco-environment monitoring. This paper investigates the accuracy and reliability of Support Vector Machine (SVM) classifier for classifying multi-spectral image of Hyderabad and its surroundings area and also compare its performance with Artificial Neural Network (ANN) classifier. In this paper, a hybrid technique which we refer to as Fuzzy Incorporated Hierarchical clustering has been proposed for clustering the multispectral satellite images into LULC sectors. The experimental results show that overall accuracies of LULC classification of the Hyderabad and its surroundings area are approximately 93.159% for SVM and 89.925% for ANN. The corresponding kappa coefficient values are 0.893 and 0.843. The classified results show that the SVM yields a very promising performance than the ANN in LULC classification of high resolution Landsat-8 satellite images.</p>

Download Full-text