Application of Object Oriented Image Classification and Markov Chain Modeling for Land Use and Land Cover Change Analysis

Abstract

Object oriented image classification (OOIC) and neural network aided Markov Chain (MC) modeling tools were used to map and predict land use and land cover (LULC) changes. A case study in the Kiskatinaw River Watershed (KRW) of Canada was presented. With an overall classification accuracy of 90.45%, the multi-temporal Landsat satellite images of KRW were analyzed for 11 selected LULC types. It was found that KRW experienced a significant wetland depletion along with a change in forest cover types from 1984 to 2010. The vulnerability of LULC change in different parts of KRW was predicted through MC modeling based on the obtained transition probability, and the results indicated slight LULC changes from 2010 with a wetland depletion of 67.89 km 2. In summary, the proposed methods generated valuable results for informed LULC management and hold the potential to be applied to other watersheds.

Multi-Layer Perceptron Neural Network and Markov Chain Based Geospatial Analysis of Land Use and Land Cover Change

“…The conversion of land use in the human subsystem driven by social activities will change land cover, and land cover changes could impact natural environment and biosphere (Rawat and Kumar, 2015;Islam, 2018a;Chen et al, 2018). As a result, land use and land cover (LULC) change resulting from anthropogenic activities has led to various concerns for environmental degradation around the globe (Islam et al, 2018b;Paul et al, 2018). The assessment of LULC change is thus of critical importance for effective environmental management and sustainable development of land resources.…”

“…The Landsat sensors have proved sensitive enough to categorize different spectral patterns related to the LU-LC classes in many complex landscape conditions (Zhao et al, 2012;Butt et al, 2015). RS analysis for change detection is usually relying on digital satellite image classification by assigning image pixels to real-world LULC feature types (Paul et al, 2018). Pixel-based classification (PBC) is a conventional method and has been broadly applied as supervised and unsupervised classification based on characteristics of single pixel (MacLean et al, 2013;Rwanga and Ndambuki, 2017).…”

“…Thus, the pixels may not represent true geographical objects when using PBC method, and its accuracy would be affected (MacLean et al, 2013;Blaschke et al, 2014). Compared with PBC, the object-based image analysis (OBIA) method generates segments from pixel-based features to produce higher classification accuracy by incorporating more information and cre-ating more recognizable segments (Frohn, 2011;Chen et al, 2013;Ma et al, 2017;Paul et al, 2018). The OBIA method also allows more types of features in the image to be exploited for classification, and it can reduce the changes resulting from the occurrence of small and spurious pixels (Toure et al, 2018).…”

“…MC model examines the stochastic nature of LULC change data and forecasts the stability of future land development. It is used by generating a transition probability matrix of LULC change from one period to another (Iacono et al, 2015;Mishra and Rai, 2016;Paul et al, 2018;Azubike et al, 2019).…”

Robust Linear Programming and Its Application to Water and Environmental Decision-Making under Uncertainty

“…It is already known that scientific management and contingency planning are the keys for addressing water-related conflicts and problems. A multitude of studies have been conducted to help decision makers to translate the sustainable development concept into operational management plans [11][12][13][14]. Mathematical programming, as one of the most useful tools that evaluate the competition scheme between sectors such as agriculture, industry and energy for limited natural resources, receives a continued interest from water, energy and environmental practitioners [15][16][17][18][19].…”