Report of Remoting sensing class project
malgasem
riyadh_city__area_final_reprot_1.docx
Final Report
“Finding the Change Detection of Riyadh using LANDSAT images”
Table of Contents
Abstract: ..................................................................................... 3
Introduction:............................................................................... 4
Data and Methods: ..................................................................... 5
Results ........................................................................................ 8
Conclusion ................................................................................ 12
Reference ................................................................................. 13
( P a g e | 15 )
Abstract:
Change detection is identified as the process of identifying the differences in the state of an object or matter or phenomenon by observing it at different times (A.Singh, 1989). In the past, there were many applications in the change detection that related to land use and land cover such as, habitat fragmentation, urban planning (D. K. Munroe, C. Croissant, and A. M., York, 2005). There are many ways to study and analyze the change of an area over time. One way that can be very helpful is to work with two processing which are Geographic Information System (GIS) and the remote sensed data. GIS are able to provide many ways to analyze the land use over an area. Also, the remote sensed data are very useful to show how the land cover changed over a period of time (J.Chilar, 2000).
2004). Change detection and monitoring by remote sensing involves the use of several multi-date images to evaluate the differences occurring in LULC between the acquisition dates of images that are due to various environmental conditions and human actions (Singh, 1989). The successful use of satellite remote sensing for LULC change detection depends upon an adequate understanding of landscape features, imaging systems, and methodology employed in relation to the aim of analysis (Yang & Lo, 2002).
Introduction:
This project is to determine the change detection over Riyadh in the past 30 years. Riyadh which is located on Latitude 24o 44’ 57” N and Longitude 46o 54’ 10”E has an area equal about 1,435 km2 with a population about 5.25 million1. Riyadh city is the capital of Saudi Arabia experiencing the horizontal expansion over the past years. That is caused by many reasons, but the most important reason is that the Saudis do not like to live in apartments, and this made many difficulties for most of them. That is because many of the services that the government provide for them is in the center of the city which makes many problem for the people who live at the edges. As known, Riyadh and other cities in Saudi Arabia does not have a public transportation, such as buses or metros which would help the people to transport easily over the city.
The motivation from this project is to use the remote sensed data to analyze the problem that could happen in Riyadh regarding this huge expansion in its area over the past years. This expansion makes many difficulty for the people who live there regarding their wat to transport from one location to another using their own cars. That is because the public transportation in that city is not available.
The data should have some preprocessing technique such as radiometric correction and atmospheric correction because they may cause variation in the data which will cause errors in the change detection. (Wang, 2010).
There are many ways to do the change detection over an area. One of these methods is by classifying the of the satellite data for land cover/land use. Then, analyzing the change detection for land cover/land use types. The analysis of change
detection comprises many methods used in order to identify, describe, and quantify
1 High Commission for the Development of Riyadh, http://www.ada.gov.sa/ada_e
the differences between some scenes at different time. The classification for the land cover/land use for the city can be classified in two which are urban area and desert.
Data and Methods:
Data were acquired from different Satellite because it will cover a big time interval. All the images were taken from Earth Explorer, USGS. The first data set which includes three images from Landsat 5-TM are taken for different years. The three images that have been chosen were acquired on March, 1986, January, 1992, and February, 2000. The other data set was acquired in 2015 using Landsat 8-OLI. The months that image has been taken was October, 2015. The images were chosen based on the quality of the images, images with lower cloud covering, and images with lower dust covering because Riyadh city is in the middle of the desert, so some images effected with the loose sand which makes dust in the air. So, the total number of images that is going to be used in this project is five images which will show the change detection in the Land-use/Land-cover for the city from 1986 to 2015. To combine the data from individual bands into a single image, Layer stacking in ERDAS was used. The following figures showing the whole scene images that acquired for the LANDSAT, and all images showed in natural color combinations.
Figure 1: March 1986 Figure 2: January 1992
Figure 3: February 2000 Figure 4: October 2015
First, all images need layer stacking method in ERDAS to combine all bands in one image. For the images that acquired using LANDSAT 4 and 5 TM, there are
6 (one thermal band) bands that were stacking together in one images (without TIR band), while for the one image acquired by LANDSAT 8 OLI, they have 9 bands (two thermal bands).
Secondly, the images, which has been layer stacked, have been subset to contain only the city instead of the whole scene. The following figures showing the subset images.
Figure 5: Subset (1986) Figure 6: Subset (1992)
Figure 7: Subset (2000) Figure 8: Subset (2015)
Third, the Radiometric Enhancement has been done using ERDAS, and the method that has been chosen was Histogram Equalization because it shows the urban better than the other methods. The following images show the Histogram Equalization Radiometric Enhancement.
Figure 9: 1986 Figure 10: 1992
Figure 11: 2000 Figure 12: 2015
Fourth, the atmospheric correction has done on the images using ENVI software. After that, the unsupervised classification has been chosen because there were no training data for the place. The images have been classified into four classes which are urban, white lands (unbuilt areas), sand, and desert. The following figures show the unsupervised classification of the images and the legend of the colors assigned to each class.
Results
After all kind of these corrections on the images, many types and tools of change detection have been used. The tools are “Delta cue”, “Discrimination function” and “Image Difference”. Each tool has its way to calculate the change over two images, and to know each type, we have to know the classification of the change detection. The delta cue and discrimination function tools are pixel-based approach. That will make them hard to use with different types of satellite images. So, for the images from different satellite, we have to classify them first. Then, we can use either discrimination function of image difference to find the change over the area at a certain period. In this report, the method of doing change detection will be image difference for the images from same satellite while the discrimination function will be used with images from different satellite. The following figures are showing the change detection over the 30 years in Riyadh.
Figure 13: The difference between 1986-1992
Figure 14: The difference between 1986-2000
From figure 13, it is clear that the city had not a significant change in that period. By looking in figure 14, we can see that the city has a dramatic expansion over the period of 1986 to 2000.
For the two images (2000 & 2015), we have to do the classification first. So, unsupervised classification has been done on both images. Figure 15 & figure 16 show the classification of the two images where the images have been classified into
4 classes. The classes are urban, Whiteland (unbuilt areas), desert, and roads.
( U RB A N R OAD S W H I TEL A ND S DE S E RT )
Figure 15: unsupervised classification (2000)
Figure 16: unsupervised classification (2015)
After that, using the discrimination function, we will get three files. The subtractive image showing the area that was in before image while it is not in the after image, and that is shown in figure 17. Additive image shows the areas that it is in the after image while it is not in the before image, and that is shown in figure 18.
So, using the layer stacking tool to put the subtractive in the red gun while the blue and green guns will have the additive. Figure 19 shows the result of the layer stacking. The bright area in figure 19 shows the areas that adding over the period from 2000 to 2015.
Figure 17: subtractive Figure 18: additive
Figure 19: the result
Conclusion
From the result, it is clearly that the city has been expanded dramatically over the past 30 years. That made a lot of problem for the people who live their regarding that the city does not has any kind of public transportation. So, there are many solutions for that problem, such as building the metros and subways. Also, having a good network of buses across the city. Another solution for the problem is by enforcing the people to live in apartments. Since it is known that the people in Saudi Arabia like to live in houses that has an area not less than 500 m2, the government could enforce the contractors to build complexes that have many apartments with wide areas.
Reference
1. A. Singh, “Digital Change Detection Techniques Using Remotely Sensed
Data,” International Journal of Remote Sensing, 989-1003 (1989).
2. M.Balakeristanan and Md A. Md Said, “Land Use Land Cover Change Detection Using Remote Sensing Application for Land Sustainability”, AIP Conference Proceeding, 1482, 425 (2012); DOI: 10.1063/1.4757507.
3. D. K. Munroe, C. Croissant, and A. M., York, “Land Use Policy and Landscape Fragmentation in an Urbanizing Region: Assessing the Impact of Zoning,” Applied Geography 25(2), 121-141 (2005).
4. H. Nagendra, S. Pareeth, and R. Ghate, “People within Parks- Forest Villages, Land-Cover Change and Landscape Fragmentation in the Tadoba Andhari Tiger Reserve, India,” Applied Geography 26, 96-112 (2006).
5. P. Hurskainen and P. Pellikka, “Change Detection on Informal Settlements Using Multi-temporal Aerial Photographs- The Case of VOI, Se- Kenya,” (2005).
6. J. Chilar, “Land cover mapping of large areas from satellites: status and research priorities,” International Journal of Remote Sensing 21(67), 1093–
1114 (2000).
7. Zhaohua Chen & Jinfei Wang, “Land use and land cover change detection using satellite remote sensing techniques in the mountainous Three Gorges Area, China”, International Journal of Remote Sensing, 31:6,15191542, DOI:10.1080/01431160903475381.(2010)
8. Anderson, “A Land Use and Land Cover Classification System for Use with Remote Sensor Data,” Geological Survey Professional Paper, No. 964, U.S. Government Printing Office, Washington, D.C., 1976, p.28.
9. R. Bunce, C. Barr, R. Clarke, D. Howard and A. Lane, “Land classification of
Great Britain,” Journal of Biogeography 23, 625–634 (1996).
10. J. Chilar, “Land cover mapping of large areas from satellites: status and research priorities,” International Journal of Remote Sensing 21(67), 1093
1114 (2000).
11.W. B. Meyer, “Past and Present Land-use and Land cover in the U.S.A.
Consequences,” 24-33 (1995).
12.O.R. Abd El-Kawy, J.K. Rød, H.A. Ismail, A.S. Suliman, “Land use and land cover change detection in the western Nile delta of Egypt using remote sensing data”, Applied Geography Journal, 483-494 (2011).
13.Das, T. “Land use/land cover change detection: an object oriented approach”,
Institute of Geoinformatics, University of Münster.(2009)
14.Chen, X., Vierling, L., & Deering, D. “A simple and effective radiometric correction method to improve landscape change detection across sensors and across time”. Remote Sensing of Environment, 98(1), 63-79.
15.Yang, X., & Lo, C., “Using a time series of satellite imagery to detect land use and land cover changes in the Atlanta, Georgia metropolitan area”. International Journal of Remote Sensing, 23(9), 1775-1798 (2002)
