Implementation of Fuzzy Logic for the High-Resolution Remote Sensing Images with Improved Accuracy

IOSR Journal of Electrical and Electronics Engineering (IOSR-JEEE)
e-ISSN: 2278-1676,p-ISSN: 2320-3331, Volume 5, Issue 3 (Mar. - Apr. 2013), PP 13-17
www.iosrjournals.org
www.iosrjournals.org 13 | Page
Implementation of Fuzzy Logic for the High-Resolution Remote
Sensing Images with Improved Accuracy
M.Priyadharshini1
, R.Karthi2
,S.N.Sangeethaa3
,R.Premalatha4
,K.S.Tamilselvan5
1
PG Scholar., Dept. of Electronics and Communication Engineering, Velalar College of Engineering and
Technology., 2
PGScholar., Dept. of Electronics and Communication Engineering, Velalar College of
Engineering and Technology., 3
Asst.Professor., Dept. of Electronics and Communication Engineering., Velalar
College of Engineering and Technology.
Abstract: In this project, an object-based approach for urban land covers classification from high resolution
satellite image using Fuzzy logic. OBIA analysis of image segmentation that is, clustering of pixels into
homogenous objects, and subsequent classification or labelling of the objects, and modelling based on the
characteristics of objects is done using MATLAB GUI model. When applied to a satellite image, the fuzzy logic
approach involves three steps. First, a membership function is computed for each pixel and each class. This
membership function expresses the likelihood that the measured radiance belongs to a class with a known
reflectance distribution. Thus, for each pixel, class memberships are assigned to the predetermined classes on
the basis of the derived membership functions. Second, three variables are retrieved from each of the class-
specific algorithms for which the pixel has membership. Third, the class memberships are used to weight the
class-specific retrievals to obtain a final blended retrieval for each pixel. The algorithm was implemented in
MATLAB GUI model and was tested on remotely sensed images of different sensors, resolutions and complexity
levels.
Keywords - GUI (graphic user interface), object based approach, high resolution satellite image.
I. Introduction
Image classification, which can be defined as identification of objects in a scene captured by a vision
system, is one of the important tasks of a robotic system. On the one side, the procedure of accurate object
identification is known to be more difficult for computers than for people. On the other side, recently developed
image acquisition systems (for instance, radar, lidar, and hyper spectral imaging technologies) capture more data
from the image scene than a human vision system. Therefore, efficient processing systems must be developed in
order to use these data for accurate image classification.
A Fuzzy system is a fuzzy logic based system. In this system fuzzy logic can be a basis for the
representation of the various kinds of knowledge or it can model the interactions and relationships among the
system variables. Fuzzy logic provides innovative tools to handle the complex and ill-defined systems where
classical tools become unsuccessful. Fuzzy systems are universal approximates of non linear functions. Two
aspects are important in fuzzy system one generating the best rule set and second tuning the membership
functions. These should relate properly the independent and dependent variable.
Color segmentation of image is a crucial operation in image analysis and in much compute vision,
image interpretation, and pattern and industrial fields such as medicine, Remote Sensing, Microscopy, content
based image and video retrieval, document analysis, industrial automation and quality control. The performance
of color segmentation may significantly affect the quality of an image understanding system. The most common
features used in image segmentation include texture, shape, grey level intensity, and color. Probabilistic latent
semantic analysis
(PLSA) [2] and latent Dirichlet allocation (LDA) [3], which were originally developed for topic
discovery in text domain, are competent to accomplish this kind of work. According to the topic models,
features are modeled as “visual words” by vector quantization, and images are regarded as documents and
modeled as mixture of latent topics. The classification or segmentation results derived from them can be more
reliant on content coherence. In addition, the effective computation based on approximate inference methods
can map the high dimensional feature counts into low-dimensional topic vectors, which makes topic models
remarkable. It has shown that representing images by topic mixtures outperforms using low-level features [4]. It
has been reported that the annotation performance of large satellite images can benefit from topic models [5].
Despite the successful application and impressive performance of topic models, they suffer from poor
spatial coherence because of the independence assumption of visual words images. A random field model such
as the Markov random field (MRF) has been employed for solving this problem since the spatial information
between neighboring regions in an image is thereby enforced. In [6] and [7], the authors have defined an MRF
over hidden topic assignment, which has been obtained by PLSA, to describe the spatial relationship of latent

Implementation Of Fuzzy Logic For The High-Resolution Remote Sensing Images With Improved
topic,the experimental results of supervised and weakly supervised settings demonstrate that the two combined
models are complementary and the segmentation and recognition accuracy is obviously improved. However, the
aforementioned method was previously proposed for supervised image classification. Furthermore, the number
of classes, which has a significant influence over classification result, is usually specified based on a prior or
educated assignment in most classification algorithms. Moreover, because of the high resolution, satellite
images have relatively richer information.
The image semantics of large structure cannot be represented properly by only using topic models. In
this letter, we extend the combination of topic models and MRFs to the topic of unsupervised satellite image
classification, more precisely, employing an MRF prior over latent topic labels, which are obtained by LDA, to
enhance the spatial coherence information. Furthermore, the label cost mentioned in [8], which has an excellent
generalization and can give its own contribution to image segmentation, is added to the random fields of latent
topics. Automatically segmenting an image into coherent parts is always an important issue in segmentation.
The minimum description length criterion was first proposed in[9] for unsupervised segmentation, to represent
the image more compactly.
In [8], the authors have pointed out that using α expansion can be more powerful, because the
segmentation based on such an algorithm relies on contour evolution and explicit merging of adjacent regions.
Based on label cost and Bayesian information criterion (BIC), we introduce an iterative algorithm over latent
topics, through which the number of classes is eventually appropriately converged. The whole process works
automatically, instead of assuming it beforehand as a constant. Meanwhile, because of the global coupling
function of label cost, both the consistency of semantics is well kept and the over smooth effect is avoided.
Moreover, the efficiency of semantics extraction is also thereby demonstrated. The rest of this letter is organized
as follows. In Section II, we review previous and related works. In Section III, we describe our approach in
detail. In Section IV, we describe our result and discussion. Finally, future work in Section V.
II. Previous Work
A. LDA-MRF MODEL
LDA-MRF combination model (latent Dirichlet allocation-Markov Random field) is proposed to obtain accurate
semantic segmentation for high resolution satellite image.
B. ENERGY OPTIMIZATION
Combining topic models with MRF is able to compensate the loss of contextual information. Local
coupling is not able to keep the consistency of semantic by large structure in satellite images, because the
influence on value of smooth factor in MRF leads to either over segment or over smooth.In order to balance
such problem, the α-expansion algorithm has been extended to incorporate label costs at each expansion; by
doing so, the energy has been represented in the form of the sum of three terms.
. (1)
Smooth cost = (2)
Data cost = (3)
Label cost = (4)
Here, the indicator function is defined on label subset L.
(5)
This extended algorithm can simultaneously optimize “smooth cost” and “label cost,” from which the
approximate solution can be kept by a certain level. The authors in [7] have discussed that some information
criteria have been employed by selection of statistical model, in order to avoid over fitting and determine the
order of the model through several iterations. Since the information criteria prefer to explain the data with fewer
and simpler models, they penalize the overly complex models [9].
The label costs could be scaled in some proportion (e.g., linear) to the estimated number of
observations per model under the suggestion of BIC. The number of models can be determined under this
algorithm by examining iterative convergence. In previous approach, label cost is added into label optimization
of the latent topic fields. At the same time, combining label cost with BIC, the number of segments in the image
is automatically determined by iterative operation. In this way, our method not only can add the global
information into the topic model but also can automatically obtain the number of segments.
The process of iterations is as follows

1) Using MRF spatial prior to constrain the latent topics, which are obtained by LDA, while setting the
initialized number of latent topics (segments).
2) Optimizing the aforementioned latent topic fields with label cost. During the optimization, figure out whether
the resulting fields match with BIC according to (1). If the minimum is reached, go to step 3); if not, i.e., the
number of latent topics (segments) is still reducing, then the iteration is continued.
3) Employing the resulting number of latent topics (segments), which is finally determined in step 2), and then
using the energy minimization algorithm to infer eventual labels.
III. Our Approach
A. DATA SELECTION
The image chosen in this project is a high resolution satellite image. The image contains three layers
(RGB). It has covered a large area.
B. OBJECT-ORIENTED CLASSIFICATION (FUZZY BASED)
1) Fuzzy Logic
Object-oriented classification is mainly based upon fuzzy logic – a mathematical technique of
quantifying uncertainty of statements. Fuzzy logic seeks to replace the two strictly logical statements “yes” and
“no” by the continuous range of values from 0 to 1, where 0 means “absolutely no” and 1 means “absolutely
yes.” Values between 0 and 1 represent various states of certainty between “no” and “yes.” This approach
emulates human thinking and takes into account even linguistic rules.
Fuzzy rules are “if – then” rules. If a condition is fulfilled, an action takes place. The concept
behind object-oriented classification is that important semantic information necessary to interpret an image is
not represented in individual pixels, but in meaningful image objects and their mutual relationships. The basic
difference, especially when compared to pixel-based procedures, is that object based classification does not
classify single pixels but rather image objects which are extracted through a segmentation procedure [12].
Basic Procedure
Object-based classification is composed of three basic procedures: image segmentation, object metric
extraction, and classification [13]. Segmentation of an image divides it into a network of homogeneous regions
at any chosen resolution. These image object primitives represent image information in an abstracted form,
serving as building blocks and information carriers for subsequent classification.
Image segmentation methods fall into two main domains: knowledge driven methods (top-down) and
data driven methods (bottom-up). In the top-down approach the user already knows what he wants to extract
from the image, but he does not know how to perform the extraction. The system tries to find the best
processing method to extract the objects by formulating a model of the desired objects. The formulated object
models give the object an implicit meaning. In the bottom-up approach the segments are generated on the basis
of a set of parameters and statistical methods for processing the whole image [12].
2) CLUSTERING ALGORITHM
There are many methods of clustering developed for a wide variety of purposes. Clustering algorithms
used for unsupervised classification of remote sensing data vary according to the efficiency with which
clustering takes place.
K means clustering
K-means is the clustering algorithm used to determine the natural spectral groupings present in a data
set. This accepts from analyst the number of clusters to be located in the data. The algorithm then arbitrarily
seeds or locates, that number of cluster centers in multidimensional measurement space. Each pixel in the image
is then assigned to the cluster whose arbitrary mean vector is closest. The procedure continues until there is no
significant change in the location of class mean vectors between successive iterations of the algorithms. As K-
means approach is iterative, it is computationally intensive and hence applied only to image subareas rather than
to full scenes and can be treated as unsupervised training areas.
K-means is one of the simplest unsupervised learning algorithms that solve the well known clustering
problem. The procedure follows a simple and easy way to classify a given data set through a certain number of
clusters (assume k clusters) fixed a priori. The main idea is to define k centroids, one for each cluster. These
centroids should be placed in a cunning way because of different location causes different result. So, the better
choice is to place them as much as possible far away from each other. The next step is to take each point
belonging to a given data set and associate it to the nearest centroids. When no point is pending, the first step is
completed and an early group age is done. At this point we need to re-calculate k new centroids as barycenters
of the clusters resulting from the previous step. After we have these k new centroids, a new binding has to be

done between the same data set points and the nearest new centroid. A loop has been generated. As a result of
this loop we may notice that the k centroids change their location step by step until no more changes are done.
In other words centroids do not move any more. Finally, this algorithm aims at minimizing an objective
function, in this case a squared error function.
The objective function,
J=∑k
j=1∑x
i=1║xi
(j)_
cj║2
Clustering is a way to separate groups of objects. K-means clustering treats each object as having a
location in space. It finds partitions such that objects within each cluster are as close to each other as possible,
and as far from objects in other clusters as possible. K-means clustering requires that you specify the number of
clusters to be partitioned and a distance metric to quantify how close two objects are to each other. Since the
color information exists in the 'a*b*' space, your objects are pixels with 'a*' and 'b*' Euclidean distance metric.
ISODATA clustering
K-mean and the ISODATA clustering algorithms are iterative procedures. In general, both of them
assign first an arbitrary initial cluster vector. The second step classifies each pixel to the closest cluster. In the
third step the new cluster mean vectors are calculated based on all the pixels in one cluster. The second and third
steps are repeated until the "change" between the iteration is small. The "change" can be defined in several
different ways, either by measuring the distances the mean cluster vector have changed from one iteration to
another or by the percentage of pixels that have changed between iterations. The ISODATA algorithm has some
further refinements by splitting and merging of clusters. Clusters are merged if either the number of members
(pixel) in a cluster is less than a certain threshold or if the centers of two clusters are closer than a certain
threshold. Clusters are split into two different clusters if the cluster standard deviation exceeds a predefined
value and the number of members (pixels) is twice the threshold for the minimum number of members.
The ISODATA algorithm is similar to the k-means algorithm with the distinct difference that the
ISODATA algorithm allows for different number of clusters while the k-means assumes that the number of
clusters is known a priori.
The objective of the k-means algorithm is to minimize the within cluster variability. The objective
function (which is to be minimized) is the sums of squares distances (errors) between each pixel and its assigned
cluster center.
Where C(x) is the mean of the cluster that pixel x is assigned to.
Minimizing the SSdistances is equivalent to minimizing the Mean Squared Error (MSE). The MSE is a
measure of the within cluster variability.
MSE=
Where N is the number of pixels, c indicates the number of clusters, and b is the number of spectral bands.
IV. Result And Discussion
The various experiment carried out in MATLAB v7.5.0 (R2007b).
TABLE I
EXPERIMENTAL RESULTS OF ENTROPY
image algorithm Entropy
Image 1
Algorithm (I) 12.44508
Algorithm (II) 7.55171
Image 2
Algorithm (I) 12.44508
Algorithm (II) 8.2281
LDA–MRF method, k-means clustering, iterative self-organizing data analysis technique (ISODATA)
clustering and Object-Oriented Classification (Fuzzy Based) are examined .On the one hand, the qualitative
evaluation is examined in terms of the semantics of the segmentation results. On the other hand, the results are
also quantitatively evaluated in terms of the purity and entropy.

V. Future Work
In future the utilization of K-means clustering and fuzzy based object-orientation classification model
for unsupervised satellite image classification will be designed so as to improve the classification accuracy.
While the above two results presented here by using isodata clustering and LDA-MRF are encouraging, there is
still a need for further improvements. Future extensions could include many algorithms related to fuzzy based
classification.
Image 1
Image 2
Image 3 Ground-Truth LDA-MRF ISODATA
Fig. 1. Classification results of three different scenes. The first column shows the input images, second column
shows the ground truth images and the third columns shows the classification results of LDA–MRF method, and
finally classification results of ISODATA clustering is displayed.
By comparing the above methods better classification results are examined.
References
[1] KanXu,Wen Yang, Gang Liu, and Hong Sun, (2013) “Unsupervised Satellite Image Classification Using Markov Field Topic
Model”.
[2] Delong. A, Osokin.A, Isack H.N, and Boykov.Y,(2010) “Fast approximate energy minimization with label costs,” in Proc. IEEE
Conf. Comput. Vis. Pattern Recog., pp. 2173–2180.
[3] Liénou.M, Maître.H, and Datcu.M, (2010), “Semantic annotation of satellite images using latent Dirichlet allocation,” IEEE Geosci.
Remote Sens. Lett, vol. 7, no. 1, pp. 28–32.
[4] Larlus.D and Jurie.F, Apr.(2009) “Latent mixture vocabularies for object categorization and segmentation,” Image Vis. Comput.,
vol. 27, no. 5, pp. 523–534
[5] MacKay.D.J.C,(2003)”Information Theory, Inference, and Learning Algorithms”, vol. 8. Cambridge, U.K.: Cambridge Univ.,p. 12.
[6] Memarsadeghi.N,Mount.D,Netanyahu.N.S, Le Moigne.J, and de Berg.M,(2007) “A fast implementation of the ISODATA
clustering algorithm,” Int. J. Comput. Geom. Appl., vol. 17, no. 1, pp. 71–103.
[7] Rosenberg and J. Hirschberg, “V-measure: A conditional entropy based external cluster evaluation measure,” in Proc. Joint
conf.EMNLPCONLL, 2007,pp. 410-420.
[8] Tang. W. Yi, H, and Chen.Y, “An object-oriented semantic clustering algorithm for high-resolution remote sensing images using
the aspect model “May (2011) IEEE Geosci. Remote Sens. Lett., vol. 8, no. 3, pp. 522–526.
[9] Verbeek.J and Triggs.B, “Region classification with Markov field aspect models,” (2007) in Proc. IEEE Conf. Comput. Vis. Pattern
Recog., pp. 1–8.
[10] Yang.W, Dai.D, Wu.J, and He.C, “Weakly supervised polarimetric SAR image classification with multi-modal Markov aspect
model,” (2010) in Proc. ISPRS, TC VII Symp. (Part B), 100 Years ISPRSAdvancing Remote Sensing Science, Vienna, Austria, Jul.
5–7, pp. 669–673.
[11] Zhu.S.C and Yuille.A.L, Sep(1996) “Region competition: Unifying, snakes, region growing, and Bayes/MDL for multiband image,
segmentation,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 18, no. 9, pp. 884–900.
[12] Civanlar,R.,Trussel,H.,(1986)Constructing membership functions using statistical data. IEEE Fuzzy Sets and Systems 18, 1 –14..
Curlander, J., Kober, W., 1992.
[13] Proc.IGARSS . IEEE Press, New York, pp. 854– 856.”Rule based system for thematic classification in SAR imagery”.

Implementation of Fuzzy Logic for the High-Resolution Remote Sensing Images with Improved Accuracy

More Related Content

What's hot (16)

Viewers also liked (20)

Similar to Implementation of Fuzzy Logic for the High-Resolution Remote Sensing Images with Improved Accuracy (20)

More from IOSR Journals (20)

Recently uploaded (20)

Implementation of Fuzzy Logic for the High-Resolution Remote Sensing Images with Improved Accuracy