VEGETATION AREA DETECTION AND CLASSIFICATION IN SATELLITE IMAGES

USING SOFT COMPUTING TECHNIQUES AND ANFIS CLASSIFIER

1
R.Shanmuga Sundaram and 2N.Santhiya Kumari
1
Asst professor/ECE, 2Professor/ECE ,Senior IEEE member
1,2
Knowledge institute of Technology, Salem,Tamilnadu,India
shanmugamme@outlook.com

Abstract

The main objective of this paper is to classify the satellite image into vegetation and non-
vegetation area. Satellite image, taken from National Satellite Meteorological Center has been
preprocessed with the colour conversion and histogram equalization process, Color conversion
technique helps to determine the accurate representation of the area presented in image with less
amount of information in the satellite image. Whereas, histogram equalization helps in increasing
the intensity level to visualize the different area details of satellite image clearly. Preprocessed
images have been processed by feature extraction technique where five different features have
been extracted with Gray level coherence matrix (GLCM) and Local Derivative Pattern (LDP)
techniques. In this proposed methodology, Adaptive Neuro Fuzzy Inference System (ANFIS)
classifier has been used to category the satellite image areas into vegetation and Non-vegetation.
Analysis was done based on the parameters such as sensitivity, specificity and accuracy. The
results obtained with sensitivity of 96.03%, 94.65 of specificity and 96.08% of accuracy shows
that better performance as compared to the existing methodologies available. It will help to
maintain climate conditions for crop cultivation, overcoming global warming. Ecological cycle
will be balanced and life span of human will be extended based on the vegetation area
availability

Keywords: Change detection, vegetation area, features, classification approach, satellite images.

1. Introduction

In India land use and land cover change are taking place at a rapid pace. Forests are the
most valuable natural resources available to the mankind on earth. On the other hand, it is also
the essential source of livelihood for the poor and marginalized sections of the society. They
provide furniture and other valuable medicine to the mankind to lead better life. Satellite images
are used to study the changes that happened in natural resources like vegetation, water bodies,
land area; mines etc., This study is useful to save the non-renewable energy resources and help to
make regulatory policies by the government. One of the major issues in India is monitoring the
forest land cover modification and is an important input for modeling ecological pattern and
environmental processes at various levels of scale. Rapid delineation of forest regions are one of
the major environmental issues facing by the world today. It has been estimated that vegetation
change threatens the worlds in terms of global warming and ecological imbalance. Vegetation
area plays a key role in terrestrial biophysical and biological process and is related to the
dynamics of global climate.Monitoring seasonal changes in vegetation activity and crop
phenology over wide areas is essential for many applications, such as estimation of net primary
production, deciding time boundary conditions for crop yield modeling and supporting decisions
about water supply. Figure.1 shows the Multispectral image with classified land use areas like
construction area, vegetation and dry land.

Figure 1 Multispectral Image with Classified Land Use Areas

Vegetation are the major part of land cover and their changes have an important influence
on the energy and mass biochemical cycles and are also a key indicator of regional ecological
environment change. Hence, detection of vegetation area in satellite images plays an essential
role in environmental changes. In this paper, satellite image with low resolution is taken and it is
enhanced with some color conversion process followed by the histogram equalization enhanced
satellite image is processed by feature extraction technique with GLCM and LDP obtained
these features are considered for the classification of vegetation and Non–vegetation area by
ANFIS classifier, The obtained results in this paper provides better results as compared to other
existing methods.

2. Related research

Masahiro Hirata et al [1,3] described the vegetation classification using Landsat TM

images which was accomplished at the level of classifying a A. herbaalba and N. mucronata
dominant community into six sub-community classes. This detailed vegetation classification was
conducted with the final aim of forage resource estimation and human impact assessment in
mind and low accuracy level achieved in this method .
Meera Gandhi.G et al. [2,4] presented an enhanced Change Detection method for the
analysis of satellite image based on normalized difference vegetation Index (NDVI) . This
employs the multi-spectral remote sensing data technique to find Vegetation Index, land cover
classification, vegetation, water bodies, open area, scrub area, hilly areas, agricultural area, thick
forest, thin forest with few band combinations of the remote sensed data. Sudhanshu et al.[5].It is
found that usage of multi-temporal landsat imageries in Detecting Seasonal Crop Abandonment
using neural network to predict the yield in season wise. Texture based segmentation [6, 8, and
11] in satellite images have considered to attain accuracy in results.
Vegetation cover [7] change detection is essential for a better understanding of the interactions
and interrelationships between humans and their ecosystem.Vegetaion area and the relation
between the soil types and it indicate which crop is suitable in particular area. They done a
community source information and secondary data and ground water resources are taken in this
work. Lot of research have been done to detect and segment the vegetation area from the non-
vegetation .analysis of the finding vegetation area in images is also very important in the field of
remote sensing .It is used for detecting green areas on Earth and detecting changes caused by
urbanization and it has classified using Support Vector Machine (SVM) classification [9, 10].
Roads, buildings and bridges are the main structural features obtained from satellite images.
Detection of clouds and shadows supports the extraction of these features [12, 26]. Different
algorithms are available for the extraction of these features, depending on the availability of
remotely sensed data.

The high resolution remote sensing data frequently available and spatial resolution
continuously bettering the object size, large details about individual objects and regions are
visible in satellite imagery. GLCHS (Grey Level Co-occurrence Hybrid Structure) method is
implemented on very high resolution satellite images to derive GLCM texture features [13]. The
high resolution satellite images are retrieved using Texture Features [14] and it uses the local
binary pattern (LBP) [15] texture feature and a block based scheme.

The various retrieval algorithms explained by Subramanian [16], they are using local
tetra patterns (LTrPs) for content-based image retrieval (CBIR). The standard local binary pattern
(LBP) and local ternary pattern (LTP) encode the relationship between the referenced pixel and
its surrounding neighbors by computing gray-level difference. The Convolutional Neural
Networks (CNNs) [17] are directly trained to produce classification maps out of the input images
and they are first initialized by using a large amount of possibly inaccurate reference data, and
then refined on a small amount of accurately labeled data.

Anita Dixit et al.[18] extracted various texture features from high resolution satellite
images. The extracted features were properly trained and classified using SVM classification
approach. Yun-Jae Choung et al. [19] developed a methodology for the detection and
classifications of water body region in satellite images using machine learning techniques. The
authors developed to construct a map model to detect and identify the water body regions in high
resolution satellite images. SVM classification [23] methodology has used by the authors for the
classification of water regions in satellite images.

Li et al.[20] ]used photo grammetry techniques to detect the particular region in high
resolution satellite images. The authors used shorelines method for improving the classification
accuracy of the satellite image object detection. Li et al. [21] applied shoreline technique to
classify various regions in satellite images. The authors constructed different shorelines for
different regions in satellite image and then the small shoreline region was identified as marked
region. Qi et al.[22] developed a methodology for satellite image classification using shoreline
processing technique. The authors used multi-task joint sparse matrix for detecting and
classifying the objects in high resolution satellite images.

Sunitha Abburu et al. [24] has explained different classification methods and analyzed
their performance with various parameters. The various classification methods like Minimum
Distance, k-Nearest Neighbour (KNN), Nearest Clustering Fuzzy C-Means (FCM) and
Maximum Likelihood (ML) Classification algorithms [25] were analyzed for best result and
maximum accuracy.
Madhavan et al.,Rajesh [27,29] described the building identification system using Adaptive
Neuro Fuzzy Inference System (ANFIS) classifier used for land cover land change analysis to
check data base information. Neurofuzzy Approach [28] has used to classify texture images with
the help of wavelet transform.

Maggiorio et al. [30] used convolutional neural networks for the detection and classification of
objects or regions in satellite images. The authors developed their algorithm with respect to two
step training technique whereas the trained features were classified into two layers. The final
layer in Convolutional neural network classification algorithm fine tunes the binary results which
were produced by various internal layers in classification algorithm.

In this work, ANFIS classification based vegetation area segmentation methodology has
been proposed. Initially, the low resolution satellite image is enhanced into high resolution
satellite image using the color conversion and it is done by converting the RGB to YUV to
separate the luminance and chrominance components It is useful for machine learning process to
represent the category of image with the minimum information.

Histogram equalization is done after color conversion to enhance the pixel intensity by
expanding the intensity level in a uniform manner. It useful to extract the important related grey
level feature and it is easily differentiate the classified area.GLCM and the Local Derivative
Pattern are applied to get the feature, which captures more detailed information related to
vegetation area easily from the enhanced image. These features are trained and used to
differentiate the pixels belonging to vegetation area and the pixels belonging to other areas using
ANFIS classifier.

3. Proposed Methodology

The flow of proposed vegetation area segmentation methodology in satellite images is

shown in Fig.2.
 Fig. 2. Proposed vegetation area segmented methodology in satellite images

3.1 Preprocessing

The objective of pre-processing is an improvement of the image data that suppresses unwanted
distortions or enhances some image features important for further processing. It is used to
conduct steps that will reduce the complexity and increase the accuracy of the applied algorithm.
The input satellite images are in Red (R) Green (G) Blue (B) format with low pixel resolution.
This low resolution RGB satellite images are converted to YC bCr format which is used to achieve
more efficient representation of input images. The following equations state the conversion of
RGB to YCbCr process.

Y  0.299 * R  0.587 * G  0.114 * B (1)

U  0.1478  R  0.289 * G  0.4368B  (2)

V  0.615 * R  0.5158  G  0.100 * B (3)

After conversion from RGB to YUV the chrominance and luminance component of the
satellite image is the luminance component is highlighted then the Chroma component
bandwidth is reduced .without any perceptual appearance this type of conversion useful in image
compression were required. Histogram equalization technique is applied in order to get the
detailed information about the required vegetation area feature of the images.

The Histogram Equalization method has been used to improve the appearance of satellite
image by adjusting intensity values without an error. Each pixel in luminance image is
individually enhanced to the higher intensity pixel value. The enhanced satellite image clearly
differentiates the vegetation area from other regions in satellite image. Fig. 3 (a) shows the input
satellite image (RGB), Fig.3 (b) shows the YCbCr converted image and Fig.3 (c) shows the
Luminance image.

(a) (b) (c)

Figure 3 (a) RGB satellite image (b) YCbCr image (c) Luminance image
 Figure 4 Enhanced satellite image

Fig.4 shows the enhanced satellite image where low resolution pixels are enhanced into
high resolution pixels.

3.2 Feature Extraction

Features correlate the linear relationship between each pixel with its surrounding pixels in
satellite images. Texture feature is the compound statistical distribution of combination of
various intensities of pixel position in the image. These Features plays an important role to
define our objective in this paper. GLCM features along with grey level feature and local
derivative pattern LDP features are used for vegetation pixel classifications in satellite images.

Gray Level Co-Occurrence ( GLCM) features

Gray level co-occurrence matrices have been used extensively in remote sensing
applications for land-cover vegetation classification and feature analysis and it is very useful in
texture analysis. It calculates the second order statistics related to image properties by
considering the spatial relationship of pixels. GLCM depicts how often different combinations of
gray levels co-occur in an image. The Spatial Relationship can be specified in different ways, the
default one is between a pixel and its immediate neighbor to its right. However it can specify this
relationship with different offsets and angles. The pixel at position (i, j) in GLCM is the sum of
the number of times the (i, j) relationship occurs in the image. The following features are
extracted from the GLCM matrix.

N
Contrast : i , j 1
pd (i  j )^ 2 (4)


N
Entropy : i , j 1
Pd (  lnPd ) (5)

(i  i )( j  j ) Pd

N
Correlation : (6)
i , j 1
ij

N
Pd
Homogeneity : 1
i , j 1 i j
(7)

Grey level features

Considering a small pixel region in luminance component satellite image, with the
described pixel at the centre, five different grey level features are extracted and these features are
taken for the classification process. These feature sets for a candidate pixel (s, t) in a sub-image J
(square sized window w*w) are given as follows,
 F1( s, t )  I ( s, t )  min{J } (8)
F 2( s, t )  max{ j}  I ( s, t ) (9)
F 3( s, t )  abs( I ( s, t )  mean{J } (10)
F 4( s, t )  std {J } (11)
F 5( s, t )  I ( s, t ) (12)

Figure.4 (a) Grey level minima feature Figure.4 (b) Grey level minima feature

Figure.4(c) Grey level minima feature Figure.4 (d) Grey level standard
deviation feature.

Figure. 4(e) Grey level pixel feature

LOCAL DERIVATIVE PATTERN FEATURES

The first order Local Binary Pattern (LBP) feature is extended to higher order feature in
four different directions as 0°, 45°, 90°, and 135° developed by Subramanian and Murala [16]. is
called as Local Derivative Pattern feature for a specific orientation is computed in n-order is
given as,

LDR ( n 1) ( rc )   0  ,45 ,90  ,135 (13)

Where, ‘k’ represents the order of the feature and it is set to 2 in this paper. ‘ is the center
pixel in 3*3 sub window.

The nth order LDP feature is computed using the following equation.
p
LDRK (rc )   2 ( p 1)  f 1( I ( K 1) ( rc ), I ( K 1) ( rp )) p  8 (14)
p 1

Where,
1, for rp and rc  0
 
f1 ( x, y)   0 else 
 
 

‘P’ represents the number of surrounding pixels and ‘p’ varies from 1 to 8.

The figure 5 shows the maximum intensity pixel feature set extracted using local
derivative pattern method.

Figure.5 Extracted LDP maximum intensity pixel feature set image

3.3 Classifications

In general, classification plays an important role in classifying each object in an image

into either class 1 or class 2. In this paper, classification role is to classify each pixel in the
satellite image into either the pixel belonging to vegetation area or the pixel belonging to non-
vegetation area. The conventional classification methods such as SVM and Neural Network did
not provide optimum pixel classification due to its non-stabilized weight function behavior of
their internal layers. In order to overcome such non-behavior functionalities, ANFIS
classification methodology is adopted in this method. This has linear weighting functional
behavior which also improves the classification rate of the vegetation pixels in satellite images.

The properties of Neural Networks and fuzzy logic integrality are combined in linear
manner which forms the closed loop as adaptive manner known as ANFIS classifier. It has five
layers as layer1 to layer 5, layer 1 act as input layer, whereas layer 5 acts as output layer. The
input layer receives the extracted features from the satellite image and the output layer produces
binary response based on the input feature set. By proper training with desired value of
vegetation pixels and non-vegetation pixel in satellite images, the system is trained as if the pixel
belong to the vegetation area the response produce by the system is high and for non-vegetation
pixel the response is low.

Figure.6(a) shows the source satellite image and Fig.6(b) shows the vegetation area
classified satellite image.

Figure. 6 (a) Source satellite image (b) Vegetation area segmented image

(a) (b) (c)

Figure. 7 (a) Source satellite images (b) Ground truth images (c) Vegetation area segmented
image by proposed method

Fig.7(a) shows the Source satellite images, Fig.7(b) shows the Ground truth images and
Fig.7(c) shows Vegetation area segmented image by proposed method.

4. Results and Discussion

In this proposed method, the satellite images are used which is taken by National Satellite
Meteorological Center (NSMC) using INSAT satellites and they are used to detect the vegetation
area. MATLAB R2016 b is used as simulation software to detect and classify the vegetation area
in the satellite images. To evaluate our methods certain standard parameters are used to analyze
the performance. The following parameter are

Sensitivity (Se) = TP/ (TP+FN) (15)

Specificity (Sp) = TN/ (TN +FP) (16)

Accuracy (Acc) = (TP+TN) / (TP+FN+TN+FP) (17)

Whereas, TP is True Positive which illustrates the correctly classified pixels in satellite
image belonging to vegetation area, TN is True Negative which illustrates the correctly classified
pixels in satellite image belonging to non-vegetation area, FP is False Positive which illustrates
the wrongly classified pixels in satellite image belonging to vegetation area and FN is False
Negative which illustrates the wrongly classified pixels in satellite image belonging to non-
vegetation area.

The relationship between all these parameters is related by a matrix which is called as
contingency matrix, which is depicted in Table 1. If the pixels belonging to vegetation area is
correlated to vegetation area in satellite image, it is defined by TP. If the pixels belonging to
vegetation area is correlated to non-vegetation area in satellite image, it is defined by FN. If the
pixels belonging to non-vegetation area is correlated to vegetation area in satellite image, it is
defined by FP. If the pixels belonging to non-vegetation area is correlated to non-vegetation area
in satellite image, it is defined by TN.

Table 1.Contingency matrix for proposed method

Regions Classified as Classified as non-

vegetation area vegetation area
Pixels belonging to TP FN
vegetation
Pixels belonging to FP TN
non-vegetation

Table 1 shows the contingency matrix performance analysis for classification on

vegetation and non-vegetation area using ANFIS classifier in terms of sensitivity, specificity and
accuracy. The proposed ANFIS based vegetation area classification method achieves 96.03% of
sensitivity, 94.65% of specificity and 96.08% of accuracy for the classification of vegetation area
pixels. The proposed non-vegetation area classification method achieves 97.10% of sensitivity,
93.28% of specificity and 95.85% of accuracy for the classification of non-vegetation area
pixels.

Table 2 Analysis of proposed vegetation and non-vegetation area using ANFIS classifier

S.No. Classified Results (%)

 Performance Vegetation area Non-vegetation
Evaluation Parameters area
1 Sensitivity 96.03 97.10
2 Specificity 94.65 93.28
3 Accuracy 96.08 95.85

Table.2 compares the proposed ANFIS based vegetation area pixel classification
methodology with Texture feature based classification with SVM described by Anita Dixit et al.
(2017) and Texture feature based classification with SVM described by Qi et al. (2017). The
proposed method achieves 96.08% of vegetation area classification accuracy while the other
state of art methods Anita Dixit et al. (2017) achieved 87.16% of classification accuracy, and Qi
et al. (2017) achieved 86.19% of classification accuracy.

Conclusion

Satellite image classification and its improved vegetation area detection method has been
proposed.Histogram equalization technique has been used to obtain the high resolution satellite
image from the low resolution image.Feature extraction has been done followed by preprossed
imagess with GLCM and LDP techniques.It has been concluded that the extracted feature
applied with ANFIS classifier yield the specificity of 96.03,sensitivity of 94.65% and accuracy
of 96.08 % provides better classifictaion as compared to the exixsting methodoligies.The
obtained results helps to identify the area and climate conditions suitable for crop cultivation.It
will acts as an alert signal for to maintain better ecological cycle in a balanced manner
.vegetation area supplioes much more oxygen air to protect our enivorenment from global
warming.Human life span will be extended based on the availability of vegetation area.

References

1. Michal Podpora, Grzegorz Paweł Korba´s, Aleksandra Kawala-Janik” YUV vs

RGB - Choosing a Color Space for Human-Machine Interaction” Federated
Conference on Computer Science and Information Systems, 29–34(2014)

2. Meera Gandhi.G, S.Parthiban, Nagaraj Thummalu, Christy. A, Vegetation Change

Detection Using Remote Sensing And Gis – A Case Study Of Vellore District,
Procedia Computer Science 57 , 1199 – 1210(2015).
3. .Abuobayda M. Shabat, Jules-Raymond Tapamo, Original A Comparative Study
Of The Use Of Local Directional Pattern For Texture-Basedinformal Settlement
Classification, Journal Of Applied Research And Technology,134(2017).

4. Masahiro Hirata , Naoki Kogab , Hitoshi Shinjo , Haruhiro ,Fujita , Gustave

Gintzburger & Akira Miyazaki, Vegetation Classification By Satellite Image
Processing In A Dry Area Of North-Eastern Syria, International Journal Of
Remote Sensing, 22:4, 507-516(2001).

5. Krystyna Michalowska 1, Ewa Glowienka 2, Beata Hejmanowska, Temporal

Satellite Images In The Process Of Automatic Efficient Detection Of Changes Of
The Baltic Sea Coastal Zone, IOP Conf. Series: Earth And Environmental Science
44 (2016)

6. Sudhanshu Sekhar Panda, Daniel P. Ames , And Suranjan Panigrahi, Application

Of Vegetation Indices For Agricultural Crop Yieldprediction Using Neural
Network Techniques, Remote Sensing ,2, 673-696;(2010)

7. Noryusdiana Mohamad Yusoff , Farrah Melissa Muharam, The Use Of Multi-

Temporal Landsat Imageries In Detecting Seasonal Crop Abandonment, Remote
Sensing ,7,11974-11991;(2015)

8. S V Sai And E V Mikhailov, Texture-Based Forest Segmentation In Satellite

a. Images, Journal Of Physics, 803 (2017).

9. Anwar Abdelrahman Aly, Abdulrasoul Mosa Al-Omran, Abdulazeam Shahwan

Sallam,Mohammad Ibrahim Al-Wabel And Mohammad Shayaa Al-Shayaa,
a. Vegetation Cover Change Detection And Assessment In Arid Environment
Using Multi-Temporal Remote Sensing Images And Ecosystem
Management Approach, Solid Earth, 7, 713–725, (2016)

10. Shuang Gao, Yuhua Peng, Huizhi Guo,Weifeng Liu, Tianxin Gao,Yuanqing Xu
11. And Xiaoying Tang,Texture Analysis And Classification Of Ultrasound Liver
Images, Bio-Medical Materials And Engineering 24 ,1209-1216(2014).

12. M. Jaganathan,A.Sabari, An Heuristic Cloud Based Segmentation Technique

Using Edge And Texture Based Two Dimensional Entropy, Cluster Computing,Jan
15 (2018).

13. Lizy Abraham, Dr.M.Sasikumar, Structural Feature Extraction From Satellite

Images, International Journal Of Scientific & Engineering Research, Volume 4,
Issue 12, December(2013)
14. Anagha S. Aghav, Prof N.S.Narkhede, Application-Oriented Approach To Texture
Feature Extraction Using Grey Level Co-Occurrence Matrix (GLCM),
International Research Journal Of Engineering And Technology, Volume: 04 Issue:
05 | May (2017)

15. Samia Bouteldja And Assia Kourgli, Retrieval Of High Resolution Satellite
Images Using Texture Features, JOURNAL OF ELECTRONIC SCIENCE AND
TECHNOLOGY, VOL. 12, NO. 2,211-215 JUNE (2014).

16. T.Vignesh, K.K.Thyagharajan, Local Binary Pattern Texture Feature For Satellite,
International Conference On Science, Engineering And Management
Research(2014)

17. Subrahmanyam Murala, R. P. Maheshwari and R. Balasubramanian, “Local Tetra

Patterns: A New Feature Descriptor for Content-Based Image Retrieval”, IEEE
Transactions on image processing, vol. 21, no. 5, (2012).

18. Maggiori E, Tarabalka Y, Charpiat G, Alliez P. Convolutional Neural Networks for Large-
Scale Remote-Sensing Image Classification. IEEE Transactions on Geoscience and
Remote Sensing. 2017 Feb;55(2):645-57.

19. Anita Dixit , Nagaratna Hedge and Dr. B. Eswar Reddy, “Texture Feature Based Satellite
Image Classification Scheme Using SVM “, International Journal of Applied Engineering
Research, Volume 12, Number 13 (2017) pp. 3996-4003.
20. Yun-Jae Choung and Myung-Hee Jo, “Comparison between a Machine-Learning-Based
Method and a Water-Index-Based Method for Shoreline Mapping Using a High-
Resolution Satellite Image Acquired in Hwado Island, South Korea,” Journal of Sensors,
vol. 2017, Article ID 8245204, 13 pages,( 2017.)

21. R. Li, K. Di, and R. Ma, “3-D shoreline extraction from IKONOS satellite
imagery,” Marine Geodesy, vol. 26, no. 1-2, pp. 107–115, (2003).

22. R. Li, K. Di, and R. Ma, “A comparative study of shoreline mapping techniques,”
in Proceedings of the 4th International Symposium on Computer Mapping and GIS for
Coastal Zone Management, Halifax, Nova Scotia, Canada, June 2001.

23. Qi, K.; Liu, W.; Yang, C.; Guan, Q.; Wu, H. Multi-Task Joint Sparse and Low-Rank
Representation for the Scene Classification of High-Resolution Remote Sensing Image.
Remote Sens. 9, 10(2017).
24. Al-doski, J., S. Mansor and H. Shafri (2013). Image classification in remote sensing.
Journal of Environment and Earth Science, (3)10: 141-148.

25. Abburu, S. and S. Golla (2015). Satellite image classification methods and techniques: A
review. International Journal of Computer Applications, 119 (8): 20-25.
26. Kanika Kalra, Anil Kumar Go swami, Rhythm Gupta, A Comparative Study Of
Supervised Image classification Algorithms For Satellite Images, International Journal Of
Electrical, Electronics And Data Communication,1(2013),10-16 .

27. Harbaš And M. Subašić, Detection Of Roadside Vegetation Using Featuresfrom

The Visible Spectrum, MIPRO,1454-1459(2007).

28. E.Madhavan,S.Kalpana, Building Identification in Satellite Images Using ANFIS

Classifier,Asian Journal of Applied Science and Technology, Volume 1,2, 283-
285,March(2017)

29. .Sushant Panigrahi,Toran Verma, Texture Image Classification using Neurofuzzy,

nternational Journal Of Engineering And Computer Science, Volume 2 Issue 7
,Page No.2309-2313, (July 2013),

30. S. Rajesh,S. Arivazhagan,K. Pratheep Moses,R. Abisekaraj, ANFIS Based Land

Cover/Land Use Mapping of LISS IV Imagery Using Optimized Wavelet Packet
Features, Indian Society of Remote Sensing 2013.

31. Afirah Taufik, Sharifah Sakinah Syed Ahmad , Land cover classification of
Landsat 8 satellite data based on Fuzzy Logic approach , IOP Conf. Series: Earth
and Environmental Science37(2016)