Scribd
Upload
180 views

Soil Classification Using Image Processing and Modified SVM Classifier

Recently the use of soil classification has gained more and more importance and recent direction in research works indicates that image classification of images for soil information is the preferred choice Various methods for image classification have been developed based on different theories or models In this study, three of these methods Maximum Likelihood classification MLC , Sub pixel classification SP and Support Vector machine SVM are used to classify a soil image into seven soil classes and the results compared MLC and SVM are hard classification methods but SP is a soft classification Hardening of soft classifications for accuracy determination leads to loss of information and the accuracy may not necessary represent the strength of class membership Therefore, in the comparison of the methods, the top 20 compositions per soil class of the SP were used instead Results from the classification, indicated that output from SP was generally poor although it performs well with soils such as forest that are homogeneous in character Of the two hard classifiers, SVM gave a better output than MLC Priyanka Dewangan | Vaibhav Dedhe "Soil Classification Using Image Processing and Modified SVM Classifier" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-2 | Issue-6 , October 2018, URL: http://www.ijtsrd.com/papers/ijtsrd18489.pdf

Uploaded by

Editor IJTSRD
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
180 views

Soil Classification Using Image Processing and Modified SVM Classifier

Recently the use of soil classification has gained more and more importance and recent direction in research works indicates that image classification of images for soil information is the preferred choice Various methods for image classification have been developed based on different theories or models In this study, three of these methods Maximum Likelihood classification MLC , Sub pixel classification SP and Support Vector machine SVM are used to classify a soil image into seven soil classes and the results compared MLC and SVM are hard classification methods but SP is a soft classification Hardening of soft classifications for accuracy determination leads to loss of information and the accuracy may not necessary represent the strength of class membership Therefore, in the comparison of the methods, the top 20 compositions per soil class of the SP were used instead Results from the classification, indicated that output from SP was generally poor although it performs well with soils such as forest that are homogeneous in character Of the two hard classifiers, SVM gave a better output than MLC Priyanka Dewangan | Vaibhav Dedhe "Soil Classification Using Image Processing and Modified SVM Classifier" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-2 | Issue-6 , October 2018, URL: http://www.ijtsrd.com/papers/ijtsrd18489.pdf

Uploaded by

Editor IJTSRD
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

International Journal of Trend in Scientific Research and Development (IJTSRD)

International Open Access Journal | www.ijtsrd.com

ISSN No: 2456 - 6470 | Volume - 2 | Issue – 6 | Sep – Oct 2018

Soil Classification
ication Using Image Processing and Modified
SVM Classifier
Priyanka Dewangan, Vaibhav Dedhe
Department of Electronics and Telecommunication Engineering,
Engineering
Shri Shankaracharya Institute of Technology & Management, Bhitauna, Uttar Pradesh,
Pradesh India

ABSTRACT
Recently the use of soil classification has gained more infrastructure, farming (agriculture) and other human
and more importance and recent direction in research activities also increases.
works indicates that image classification of images for
soil information is the preferred choice. Various Satisfying as a result, land and its associated natural
methods for image classification have been developed resources like forest, vegetation, etc are being
based on different theories or models. In this study, subjected continually to changes and these
t changes in
three of these methods Maximum Likelihood turns affect the ecosystem. Even water resources like
classification (MLC), Sub pixel classification (SP) rivers, streams and wetlands that may be found in
and Support Vector machine (SVM) are used to areas where such activities occur are also affected.
classify a soil image into seven soil classes and the For example, when changes occur in vegetation;
results compared. MLC and SVM are hard wildlife habitat, fire conditions; aesthetic
a and
classification methods but SP is a soft classification. historical values and ambient air quality, are all
Hardening of soft classifications for accuracy affected.
determination leads
eads to loss of information and the
accuracy may not necessary represent the strength of As human and natural forces are transforming the
class membership. Therefore, in the comparison of the landscape, resource agencies find it increasingly
methods, the top 20% compositions per soil class of important to monitor and assess these alterations.
the SP were used instead. Results from the Land use and soil is therefore regarded as the single
classification, indicated
dicated that output from SP was most important factor of environmental change such
generally poor although it performs well with soils as deforestation, habitat fragmentation, urbanization,
such as forest that are homogeneous in character. Of and wetland degradation. Soil deals with the physical
the two hard classifiers, SVM gave a better output features or vegetation as evident on the land whereas
than MLC. land use is about what
hat economic activity or use the
land is put to Research in land use. Soil studies have
Keywords: Soil Classification, Image Processing, generated so much interest locally and internationally
Support Vector Machine, SVM due to concerns globally on land use. Soil changes
and its consequences to the environment. It has
I. INTRODUCTION therefore becomeme one of the crucial elements in
It is an undeniable statement that ‘land is an important images classification for scientific research and real-
real
asset and a means to sustain livelihood’. It is the key life earth science applications[2].
applications One of the
resource for most human activities including forestry, fundamentals required for such studies are maps.
agriculture, industry, mining, etc. Land is therefore a Various methods are used for the production of these
fundamental factor of production closely linked with maps, however, the application of remote sensing for
the economic growth of a nation and its peoplepeople[1]. map production is increasing become the relatively
However, as the population increases, demand for cheap and quick method of acquiring up-to-date
up
land for use in settlement, construction of information over a large geographical area.
Conventional
onventional ground survey methods of mapping are

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct


Oct 2018 Page: 504
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
2456
labour intensive, time consuming and are done
relatively infrequently[3].. Especially in fast changing
environments, maps produced using ground survey
methods will soon become obsolete with passage of
time.

II. PROBLEM IDENTIFICATION


Cryptography The availability of up-to to-date land use
soil information is central to much resource
management, planning and monitoring programmes.
Maps are fundamental for the provision of such land
use – soil information and processes that enhances or
leads to the production of such current maps in a cost
cost-
effective manner is essential. Recent direction in
research works indicates that classification of satellite
images for soil information is the preferred choice for
producing such maps.

Out of the many methods for image classification,


three were chosen Maximum Likelihood
Classification, Sub-pixel
pixel Classification and Artificial
Neural Network Classification[3].. In this work, Figure 1: Fundamental Structure of soil classification
support vector machine is chosen because they process
provide a varied of options, good for comparison as;
1. Maximum Likelihood is a supervised parametric Radiometric Correction
hard classifier. It has been identified above; radiometric corrections
pixel Classification is a supe
2. Sub-pixel supervised non- are important and mandatory when using multi-date
multi
parametric soft classifier. images datasets. This must be done so that images
3. Artificial Neural Network Classification is a obtained by sensors at different times are made
supervised non-parametric
parametric hard classifier. comparable in terms of radiometric characteristics.
Techniques like image enhancement,
enhanceme normalization,
The focus of this project is to explore the capabilities calibration etc have been applied to multi-date
multi
of the selected methods (Maximum Likelihood satellite images to increase the amount of information
Classification, Sub-pixel Classification and Neural for improve interpretability. In this study however, no
Network Classification) to effectively represent soil atmospheric correction was applied because data on
types. atmospheric characteristics was not available.
Secondly only a single-date
date image is being used for
III. METHODOLOGY the classification and therefore atmospheric correction
Image Pre-Processing can be ignored.
Image pre-processing
processing are actions or processes
undertaken prior to the main data analysis and Image Processing
extraction of information. They are of two main forms In remote sensing, images are historically processed
– radiometric correction or geometric correction. digitally because of two important principal
princi areas of
Radiometric corrections are needed to correct the data application namely; the improvement of the spectral
for sensor irregularities and unwanted sensor or information to enhance the process of visual
atmospheric noise, and converting the data so they interpretation and the processing of image data for
accurately represent the reflectedd or emitted radiation computer assisted classification. The aim of both
measured by the sensor[4]. processes is to increase spectral separability
separabili of the
object features in the image.

Commonly used image enhancement techniques


include image reduction, image magnification;
transect extraction, contrast adjustments (linear and

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct


Oct 2018 Page: 505
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
2456
non-linear),
linear), band rationing, spatial filtering, Fourier
transformations, principal components analysis, and
texture transformation (Jensen, 1996). Two of these
techniques; Vegetation Indices developed under the
band rationing technique and Principal Components
Analysis will be considered briefly as they were used
in this study. (d) Humus Clay

Image classification using Support vector Machine


Support vector machine (SVM) in machine learning
are supervised models associated with respect to
learning algorithms. These are mainly used for
analysing the data for regression and classification.
For a set of training examples, it belongs to either one (e) Peat
of the two categories; a support vector machine
algorithm for training generates a model which tells
the new thing falls in to which category by a non non-
probabilistic binary classifier. SVM model is the
example
ample on depiction of points in space which is
mapped. Thus, the data of different types are
separated by as wide as possible. The new data are (f) Sandy Clay
mapped and categorized according to which part of
the group they fall on. Support Vector Regression:
Developed by Vapnik (1998).

IV. RESULTS
The image dataset used are clay, Clayey Peat, Clayey
Sand, Humus Clay, Peat, Sandy Clay, and Silty Sand. (g) Silty Sand
Out of this database two images from each are shown Figure 2: Soil classification database
in figure 2.
The algorithm proposed here may have significant
improvement over some of the other past soil
classification systems.

Binary classification: binary classifier (Table:1) is


used to govern whether the soil type is sandy or not.
(a) Clay Support vector machine (SVM) do the classification
of the non-sandy soil. The soil types are better
classified here Majority of the misclassified objects
are relayed near to the segment line. Near the segment
boundary Measurements spotted as often noisy and
thus can be decided that the enactment of the
classifiers was excellent.
(b) Clayey Peat

(c) Clayey Sand

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct


Oct 2018 Page: 506
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
2456
IV. CONCLUSION and the analysis of land use/land cover change,”
These Different techniques has been used and SVM Agric. Ecosyst. Environ.,, vol. 101, no. 2–3,
2 pp.
classifier is found to outperform over all other 111–115, Feb. 2004.
techniques.
3. H. Yuan, C. Van Der Wiele, and S. Khorram, “An
Automated Artificial Neural Network System for
REFERENCES Land Use/Land Cover Classification from Landsat
1. F. Ellis, “The Determinants of Rural Livelihood TM Imagery,” Remote Sens.,
Sens. vol. 1, no. 3, pp.
Diversification in Developing Countries,” J.
243–265, Jul. 2009.
Agric. Econ.,, vol. 51, no. 2, pp. 289
289–302, Nov.
2008. 4. T. M. Lillesand and T. M. Lillesand, Remote
sensing and image interpretation.
interpretation 2003.
2. H. Nagendra, D. K. Munroe, and J. Southworth,
“From pattern to process: landscape fragmentation

@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 6 | Sep-Oct


Oct 2018 Page: 507

You might also like