IRJET- Geological Boundary Detection for Satellite Images using AI Technique

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 04 | Apr 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 1173
Geological Boundary Detection for Satellite Images using
AI Technique
Apurva Pilay1, Chhaya Pawar2
1Dept of Computer Engineering, Datta Meghe College of Engineering, Airoli
2Assistant Professor, Dept. Of Computer Engineering, Datta Meghe College of engineering, Airoli
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract - Many image processing and analysis techniques
have been developed to aid the interpretation of remote
sensing images and to extract as much informationaspossible
from the images. The choice of specific techniques or
algorithms to use depends on the goals of each individual
project. For each application it is necessary to develop a
specific methodology to extract information from the image.
To develop a methodology it is necessary to identify a
procedure based on image processing techniques that is more
adequate to the problem solution. In spite of the application
complexity, some basic techniques are common in most of the
remote sensing applications named as image registration,
image fusion, image segmentation and classification. Hence,
proposed method aims to present the use of image processing
techniques to solve a general problem on remote sensing
application. In proposed method, we examined some
procedures commonly used in analyzing remote sensing
images by using a novel method of Particle swarm
optimization technique (PSO).
Key Words: Image Processing, Satellite images, particle
swarm optimization, 2D convolution
1. INTRODUCTION
Satellite images have many applications in meteorology,
oceanography, fishing, agriculture, biodiversity
conservation, forestry, landscape, geology, cartography,
regional planning, education, intelligence and warfare.
Images can be in visible colors and in other conducted using
specialized remote sensing software.
Satellite image processing has proven to be a powerful tool
for the monitoring of the earth’s surface to improve our
perception of our surroundings has led to unprecedented
developments in sensor and information technologies.
However, technologies for effective use of the data and for
extracting useful information fromthedata ofsatelliteimage
processing are still very limited since no single sensor
combines the optimal spectral, spatial and temporal
resolution. The conclusion of this,accordingtoliterature, the
remote sensing still lacks of software tools for effective
information extraction from Satellite image processingdata.
For many parts of the world, medium to high resolution
remote sensing satellites will only acquire data after the
satellite has been programmed to do so. In these
circumstances, coverage of the affected area is likely to be
delayed and possibly missed. However, when major
disasters unfold, most satellite operators will schedule
imagery collection, even without confirmed programming
requests, either on humanitarian grounds or in the hope of
data sales. These all the satellite image processingdrawback
effects the whole disaster mitigation process, so we are
processing the novel techniques to integrate the system by
various combination of algorithm to amalgamate the
geological boundary data with geohydrology data and
lithosphere data like HI climb mountains, terrain,
sedimentary basin, rifts etc. The proposed methodwill focus
on flood rescue and mitigation mainly. The objectives of
proposed method is to present an advanced method for
combination of multi-spectrum RGB images for multi-
spectrum image fusion. The proposedmethodcanbeusedto
analyze the maxima features in a particular image.
2. LITERATURE SURVEY
Aparna Joshi and Isha Tarte discussed damageidentification
and assessment using image processing on post disaster
satellite imagery. SLIC i.e. simple lineariterativeclusteringis
used for segmenting which is a simplemethodtodecompose
an image in visually homogeneous regionswhichisbased on
spatially localized version of k-means clustering. Random
forest algorithm is used for classification which works by
creating a set of decision trees from randomly selected
subset of training set, aggregating the votes from different
decision trees to decide the final class of the test object. This
algorithm has high accuracy results. [1]
Milad Janalipour & Mohammad Taleai concentrated on
building change detection after earthquake using multi-
criteria decision analysis based on extracted information
from high spatial resolution satellite images. Adaptive
network based fuzzy inference system is used which is a
combination of fuzzy systems and neural networks. To
address real world problems, ANFIS is extremely useful asit
addresses objective knowledge as well as subjective
knowledge i.e. knowledge including mathematical models
and design requirements. [3]
This paper by D.C. Mason, L. Giustarini,J.Garcia-Pintado, and
H.L. Cloke investigates whether urban flooding can be
detected in layover region using double scattering between
ground surface and walls of adjacent buildings. The method
estimates double scattering strengths using SAR image in
conjunction with a high resolution LiDAR height map of the

urban area. A SAR simulator is applied to the LiDAR data to
generate maps of layover and shadow and estimate the
positions of double scattering curves in the SAR image. [4]
Weixing Wang, Nan Yang, Yi Zhang, Fengping Wang, Ting
Cao, Patrik Eklund, analyse road features, road model,
existing difficulties and interference factors for road
extraction. Secondly, the principle of road extraction,
advantages and disadvantages of various methods and
research achievements are briefly highlighted. Conclusion
states that single method is not enoughtogetoptimal results
of road extraction thus various methods need to be
combined in order to be used in real applications. [5]
The method proposed by DejanVukadinov,Raka Jovanovicis
based on canny edge detection and threshold. Proposed
algorithm consists of five main steps viz. image extraction,
histogram matching, Gaussian blur filter, locally adaptive
threshold, edge detection. The system is tested using
different artificial island of Dubai coastline. Though the
images are very complex, results obtained are accurate.[10]
3. METHODOLOGY
3.1 System Architecture
Fig -3.1: Architecture of System
The system consists of the following main steps:
1. Read the source image into input.
2. For pre-processing step, the input image is
converted to BMP format from RGB format.
3. BMP format dataset is analysed into red, green and
blue plane which helps analysing each pixel
individually.
4. Histogram is generated which helps in
differentiating red,greenandblueplanefromwhich
net deterministic value for each pixel
differentiation.
5. Training parameters are obtained from histogram
differentiation which are integrated with
intelligence i.e. particle swarm optimization
algorithm.
6. Various convolution models for various planes are
generated. Then geological parameters of pixels
obtained from histogram technique are compared
with convolution results.
7. A 3D matrix is obtained from convolved results
where each dimension refer to a particular
geological boundary with pixel differentiation of
land-water, water-Greenland, and land-Greenland.
8. Non pixel data generated from convolution is
removed then it is integrated with PSO.
9. PSO decides the maxima and minimainconvolution
models supplied to it as input.
10. PSO technique integrates the similar color pattern
on a particular pixel boundary of our convolution
model. Similarly it doesforotherdimensionandour
color pattern on that particular image is generated
which is our final output.
This method is tested on before flood andafterfloodsatellite
images of Kerala obtained from NASA website. Pre-
processing is carried out which coverts image from RGB to
BMP format and differentiated red, green and blue planes. It
generates histogram and convolution models for both
images and final result. The difference between two outputs
can be seen clearly w.r.to geological boundary detection.
Results obtained are as follows:
Fig – 3.2 Input Image
Input Satellite Image
Pre-processing
Final Result
2D Convolution
Histogram Generation
Edge Detection
Particle Swarm
Optimization algorithm

The input image is the satelliteimageofKerala capturedpost
flood which occurred on August 22nd 2018. The image is in
JPEG format.
Fig- 3.3 Convolution of green and blue plane keeping
red plane as reference
The 2D convolution is obtained for green and blue plane
keeping the red plane constant. It is seen from the result of
convolution that blue color is highlighted from the
convolution of green and blue planes.
Fig- 3.4 Histogram calculation of input image
The histogram is calculated for the input image of pixel
numbering and pixel intensity. It shows which pixel has
occurred how many times.
Fig - 3.5 Edge Detection in four iterations
The edge detection algorithmisappliedoninputimagealong
with knowledge of convolution and histogram which gives
optimized results in the iterations of edges.
Fig – 3.6 Pixel accuracy and error bound graphs
The graphs of pixel accuracy and error bound are calculated
w.r.to number of iterations taken in the edge detection. The
error is between range of -1 to 1 for four iterations and pixel
accuracy is maximum for two iterations and lowers after
that.
Fig – 3.7 Input to particle swarm optimization
algorithm

The input image, edge detection result which obtainedusing
convolution, histogramandbitmapformatimagecreated out
of input image is given as input to particle swarm
optimization.
Fig – 3.8 Output Image
The output image is the output of particle swarm
optimization algorithm which shows flooded area with blue
color, land area with green color and red area is error which
is covered by clouds in red color.
Fig – 3.9 Comparison of input and output
The comparison of input and output image is shown in this
figure 3.9. The exact similarity in flooded area from input
image and blue colored area from output image can be seen
and boundaries aredetected betweenwaterandlandclearly.
4. CONCLUSION
The results obtained from proposed method will be a great
measure for predicting and analyzing impactoffloods.It will
help rescue teams to address high alert areas first so,
minimum or no loss of life will be achieved. In future, the
method can be modified to be used for coastline detection,
urbanization, deforestation and earthquakes.
REFERENCES
[1] Cartwright, S. 2005: National Civil Defense Emergency
Management Plan Order 2005. Published under the
authority of the New Zealand Government, Wellington, New
Zealand, 68 pp.
[2] Aparna Joshi, I. Tarte, S. Suresh and S. G. Koolagudi,
"Damage identification and assessment using image
processing on post-disaster satellite imagery," 2017 IEEE
Global Humanitarian Technology Conference (GHTC), San
Jose,CA,2017,pp.1-7.
doi: 10.1109/GHTC.2017.8239286
[3] Menderes, Aydan & Erener, Arzu & Sarp, Gulcan. (2015).
Automatic Detection of Damaged BuildingsafterEarthquake
Hazard by Using Remote Sensing and Information
Technologies. Procedia EarthandPlanetaryScience. 15. 257-
262. 10.1016/j.proeps.2015.08.063.
[4] Building change detection after earthquake using multi-
criteria decision analysis based on extracted information
from high spatial resolution satellite images Milad
Janalipour & Mohammad Taleai received 06 Aug 2015,
Accepted 03 Nov 2016, Published online: 12 Dec 2016
https://doi.org/10.1080/01431161.2016.1259673
[5]D.C. Mason, L. Giustarini, J. Garcia-Pintado, H.L. Cloke,
Detection of flooded urban areasinhighresolutionSynthetic
Aperture Radar images using double scattering,
International Journal of Applied Earth Observation and
Geoinformation, Volume 28, 2014, Pages 150-159, ISSN
0303-2434, https://doi.org/10.1016/j.jag.2013.12.002.
[6] Weixing Wang, Nan Yang, Yi Zhang, Fengping Wang,Ting
Cao, Patrik Eklund, A review of road extraction from remote
sensing images, Journal of Traffic and Transportation
Engineering (English Edition), Volume 3, Issue 3, 2016,
Pages 271-282, ISSN 2095-7564
[7]José Rouco, Catarina Carvalho, Ana Domingues, Elsa
Azevedo Campilho, A robust anisotropic edge detection
method for carotid ultrasound image processing ,Procedia
Computer Science, Volume 126, 2018, Pages 723-732, ISSN
1877-0509
[8] Journal of Computational Design and Engineering 3
(2016) 191–197 Depth edge detection by image-based
smoothing and morphological operations Syed Mohammad
Abid Hasan, KwangheeKonSchool ofMechatronics,Gwangju
Institute of Science and Technology, Gwangju, South Korea
Received 3 December 2015; received in revised form 5
February 2016; accepted 10 February 2016
[9] Journal of Chemical and Pharmaceutical Sciences ISSN:
0974-2115 JCHPS Special Issue 1: February 2017
www.jchps.com Page 96 Identification of Satellite Image by
Using DP Clustering Algorithm for Image Segmentation and
Clustering S. Sowmiya*, S. P. Yazhini Department of

Computer Science and Engineering, M. KumarasamyCollege
of Engineering, Karur, Tamil Nadu.
[10] Thilagamani, S. (2011). Object Recognition Based on
Image Segmentation and Clustering. Journal of Computer
Science. 7. 1741-1748. 10.3844/jcssp.2011.1741.1748.
[11] International Journal of Computers
http://www.iaras.org/iaras/journals/ijc
"An Algorithm for Coastline Extraction from Satellite
Imagery" DEJAN VUKADINOV
RAKA JOVANOVIC MILAN TUBA 2017
[12] Chandrani Singh, M. Shekar, Arun Singh, and R. K.
Chadha"Seismic Attenuation Characteristics along the Hi-
CLIMB Profile in Tibet from Lg Q Inversion",Bulletin of the
Seismological Society of America, Vol. 102, No. 2, pp. 783–
789, April 2012, doi: 10.1785/0120110145
[13] Karen E. Joyce,1* Stella E. Belliss,2 Sergey V.
Samsonov,1 Stephen J. McNeill2 and Phil J. Glassey1, "A
review of the status of satellite remote sensing and image
processing techniques for mapping natural hazards and
disasters", Progress in Physical Geography 33(2) (2009) pp.
183–207 DOI: 10.1177/0309133309339563.
[14] Taher M. Sodagar and Don C. Lawton,"Time-lapse
seismic modelling of CO2 fluid substitution in the Devonian
Redwater Reef, Alberta, Canada" Geophysical Prospecting
doi: 10.1111/1365-2478.12100
[15] Umbaugh, Scott E (2010). Digital image processing and
analysis: human and computer vision applications with
CVIPtools (2nd ed.). Boca Raton, FL: CRC Press. ISBN 978-1-
4398-0205-2
[16] Mehmet Sezgin and Bulent Sankur, Survey over image
thresholding techniques and quantitative performance
evaluation, Journal of Electronic Imaging 13(1), 146–165
(January 2004). doi:10.1117/1.1631315
[17] Zhang, Y. (2011). "Optimal multi-level Thresholding
based on Maximum Tsallis Entropy via an Artificial Bee
Colony Approach". Entropy. 13 (4): 841–859.
doi:10.3390/e13040841.
[18] Kennedy, J.; Eberhart, R. (1995). "Particle Swarm
Optimization". Proceedings ofIEEEInternational Conference
on Neural Networks. IV. pp. 1942–1948.
doi:10.1109/ICNN.1995.488968.
[19] Kennedy, J. (1997). "The particle swarm: social
adaptation of knowledge". ProceedingsofIEEEInternational
Conference on Evolutionary Computation. pp. 303–308.
[20] Mohd Afizi Mohd Shukran, Yuk Ying Chung, Wei-Chang
Yeh, Noorhaniza Wahid and Ahmad Mujahid Ahmad Zaidi.
2011. Image ClassificationTechniqueusingModifiedParticle
Swarm Optimization
[21] Shi, Y.; Eberhart, R.C. (1998). "A modified particle
swarm optimizer". Proceedings of IEEE International
Conference on Evolutionary Computation. pp. 69–73.
[22] Braik, Malik & Sheta, Alaa & Ayesh, Aladdin. (2007).
Image Enhancement Using Particle Swarm Optimization,
Proceedings of the world congress on engineering 2007 Vol
1,WCE 2007,July 2-4,2007, London, U.K.

IRJET- Geological Boundary Detection for Satellite Images using AI Technique

Related slideshows

More Related Content

IRJET- Geological Boundary Detection for Satellite Images using AI Technique