Cramer’s V Test Discretization Based Spatial Decision Tree Learning for Land Cover Classification
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2015, Vol 3, Issue 12
Abstract
Given learning samples from a raster data set for spatial data mining, spatial decision tree learning models is used to estimate the decision tree classifier that minimizes classification errors as well as salt-and-pepper noise. The problem has important societal applications such as land cover classification for natural resource management. However, the problem is challenging due to the fact that learning samples show spatial autocorrelation in class labels, instead of being independently identically distributed. Related work relies on local tests (i.e., testing feature information of a location) and cannot adequately model the spatial autocorrelation effect, resulting in salt-and-pepper noise. In similarity, we proposed a Cramer’s V Test Discretization for feature selection in the land covers in which the images with noise can be reduced with morphological filter and extract the more feature descriptor for land cover classification with spatial auto correlation property for discrete features. Preliminary results showed that Cramer’s V Test reduces classification errors and salt-and-pepper noise. This paper extends our recent work by introducing a new test approach with adaptive neighbourhoods that avoids over-smoothing in wedge-shaped areas. Cramer's V-based discretization (CVD) algorithm is proposed to optimally partition the continuous features into discrete ones. Two association-based feature selection indexes has integrated for spatial autocorrelation, the CVD-based association index (CVDAI) and the class-attribution interdependence maximization (CAIM)- based association index (CAIMAI), derived from the CV-test value, are then proposed to select the optimal feature subset. Experiment results on real world data sets show that proposed technique improves classification accuracy, and that our computational refinement significantly reduces training time.
Authors and Affiliations
Miss. S. Joy Princy, Mrs. S. Subadra
Keywords
Related Articles
- EP ID EP21445
- DOI -
- Views 285
- Downloads 4
Design and Implementation of Area Efficient BPSK and QPSK Modulators Based On FPGA
Digital communications devices designed on FPGAs are capable of affording multiple communications protocols without the need to arrange new hardware, and can support new protocols in a matter of seconds. In addition, FP...
Improvement of Bounce Characteristics on Locally Available Soil to Prepare Cricket Pitches
Cricket is one of the most popular sport in the world. Games progresses on hard and flat surface known as pitch. A model of pitch was developed with soil and aggregate to scrutinize their behaviours. The physic of pitch...
Residential Building Construction
When more than half of the floor area is used for dwelling purposes the building is regarded as residential building. Single-Family house, Condominiums, Townhouses, Cooperatives, Multi-family house, land these are six t...
Security Analysis of Distributed Denial of Service Attacks Using Machine Learning a Survey
Denial of service attack is the major problem in cyber network. DoS/DDoS attacks make the system performance slow. This paper discusses about DoS/DDoS attack detection algorithm and its various versions used for providi...
A Review on Multiplier Circuits based on Various Performance Parameters
A processor consumes most of its hardware resource and time in carrying out multiplication operation than all other arithmetic operations like addition and subtraction. The central processing unit in a computer devotes...