Manual of Remote Sensing
1 like1,300 views
The document is a comprehensive manual on remote sensing, detailing the processes involved in satellite image download, image preprocessing, interpretation, and map production. It includes step-by-step instructions for tasks such as layer stacking, metadata analysis, and various classification techniques, both supervised and unsupervised. Specific software tools and methods are discussed for enhancing image quality and interpreting geographical data effectively.
Manual of Remote Sensing
- 2. 1 | P a g e SUBMITTED BY SUBMITTED TO RAFIUL ALAM RONY BASAK REG NO:2013135006 ASSISTANT PROFESSOR & MD.TARIQUL ISLAM LECTURER DEPT. OF GEE DATE: 18.2.2018
- 3. 2 | P a g e CONTENT SERIAL NO NAME OF CHAPTER PAGE NO 1 CHAPTER ONE (SATELLITE IMAGE DOWNLOAD) 3-5 2 CHAPTER TWO (LAYER STACK) 6-7 3 CHAPTER THREE (BASIC FUNCTION) 8-10 4 CHAPTER FOUR (MULTISPECTRAL BAND COMBINATION& IMAGE ENHANCEMENT) 11-12 5 CHAPTER FIVE(SUBSET) 13-18 6 CHAPTER SIX (IMAGE PREPROCESSING) 19-24 7 CHAPTER SEVEN (IMAGE INTERPRETATION) 25-32 8 CHAPTER EIGHT (MAP PRODUCTION 33-40
- 4. 3 | P a g e CHAPTER ONE (SATELLITE IMAGE DOWNLOAD) For downloading satellite image go to https://earthexplorer.usgs.gov. If you don’t have an account then sign up for opening an account with giving the actual information. Search criteria>coordinate> Add coordinate in two format> degree/minute/second or decimal
- 5. 4 | P a g e Or you can also use map by cursor,zoom in to the desire area and add place mark by clicking mouse. After adding coordinate go to data range>>select the date range that you want. Then select dataset>Landsat>select which Landsat image you want, check the option>result
- 6. 5 | P a g e After showing result you search the date of image that you want and also see various information about image such as footprint, raw data, band information, preview on the exact location. etc. After that there is a download option in the tray and download the TIFF format image for getting all band.
- 7. 6 | P a g e CHAPTER TWO (LAYER STACK) Here the TIFF format image. (LT05_L1TP_136045_20091206__01_T1.tar) means LT05-Landsat 5 L1TP is sensor name, information And there is a date that the image is captured and the band number also given. Go to raster>>spectral>>layer stack, Input the individual file one by one and locate the output file.
- 8. 7 | P a g e file>>open>>raster layer>>open the destination image Go to content>>right click the stacked image>>fit layer to window.(A) This is the final output which is layer stacked image.(B) (A) (B)
- 9. 8 | P a g e CHAPTER THREE (BASIC FUNCTION) Add new viewer Home>>add views>>create new 2d views. Link Viewers Open new image in 2nd viewer Home>>link viewers>>link views>>control zoom in one viewer and show the exact location in another viewer.
- 10. 9 | P a g e Metadata Home>select the image>metadata>projection>see the projection of the image Projection is a method by which the curved surface of the earth is portrayed on a flat surface. This generally requires a systematic mathematical transformation of the earth's graticule of lines of longitude and latitude onto a plane. If you need re projection , then raster>>re project >>select the categories such as WGS 84,north,UTM 46.
- 11. 10 | P a g e Raster Option Right click on mouse>>Open raster layer>>select image>>raster option>change the band combination Add new viewer>>add raster layer>>open output file
- 12. 11 | P a g e CHAPTER FOUR (MULTISPECTRAL BAND COMBINATION& IMAGE ENHANCEMENT) Go to multispectral and change sensor,band combination such as true color 3,2,1 and false color 4,3,2. Manually one can change the combination or use stablished functional system. Go to adjust radiometry and changes various stretches.
- 13. 12 | P a g e Go to breakpoint and change them for better quality Change Discrete DRA, contrast and also brightness
- 14. 13 | P a g e CHAPTER FIVE(SUBSET) Shape file method Right click>open vector layer(a) Select the shape file>open it(b) (A) (B) Select the shape file>vector dialog>re projection Re project the shape file as same as the layer stacked image projection
- 15. 14 | P a g e Open the projected shape file>select the shape file>home>copy(a)>paste(b) (a) (b) After it there seems a AOI file automatically add in the contents>>right click>>save layer as>>save the AOI file
- 16. 15 | P a g e Then raster>>subset and chip>>create subset image>>input the image and select output location Select AOI file from the destination folder Then open raster layer>>open the subset image
- 17. 16 | P a g e Drawing Method Drawing>>insert geometry>> select polygon Draw polygon on the layer stacked image>>then automatically add a AOI file in contents>>save layer as>>save the AOI fi
- 18. 17 | P a g e Raster>>subset & chip>>create subset image>>select image>>select output location>>select AOI file This is the subset image
- 19. 18 | P a g e Subset chip method Multispectral>>subset & chip>>place the box in exact location>>select location of the output This is the final output
- 20. 19 | P a g e CHAPTER SIX (IMAGE PREPROCESSING) Correction: Image processing is a process which makes an image interpretable for a specific use. There are many methods, but only the most common will be presented here. Radiometric Correction Procedures that correct or calibrate aberrations in data values due to specific distortions from such things as noise reduction or instrumentation errors (such as striping) in remotely sensed data. Noise reduction Applies an edge-preserving smoothing techniques. Raster>>Radiometric>>noise reduction>>input file>>select the output location correction radiometric Atmospheric Geometric
- 21. 20 | P a g e Open raster layer>>open corrected image>>compare Brightness Inversion This dialog allows both linear and non linear of the image intensity range.Dark details become light and light details become dark. Raster>>radiometric>>brightness inversion>>input and output file in selected location
- 22. 21 | P a g e Open raster layer>>corrected photo Histogram Equalization Apply a nonlinear contrast stretch that redistributes pixel values so that there are approximately the same number of pixels with each value within a range. Raster>>radiometric >>histogram equalization>>input and output file and location. Open raster layer>>open corrected image>> go to metadata>>histogram Go to metadata of previous layer- stacked image>>histogram Then compare the two images
- 23. 22 | P a g e (B) Here the corrected image histogram show the largely distributed pixel value . Atmospheric Correction Haze reduction Atmospheric effects can cause imagery to have a limited dynamic range, appearing as haziness or reduced contrast. Use this dialog to sharpen an image using Tasseled Cap or Point Spread Convolution. For multispectral images, this method is based on the Tasseled Cap transformation that yields a component that correlates with haze. This component is removed and the image is transformed back into RGB space. For panchromatic images, an inverse point spread convolution is used. Radiometric>>haze reduction>>input and output the file >>open new raster layer(a) open corrected file(b) (A)
- 24. 23 | P a g e (a) (b) Compare and see the changes Geometric Correction Mosaic Use MosaicPro workstation to join georeferenced images and form a larger image or a set of images (these mosaicked project files are named with an .mop file extension). The input images must all contain map and projection information and have the same number of layers. They do not need to be in the same projection or have the same pixel cell sizes. Calibrated input images are also supported. Go to raster>>Mosaic pro>>display add images dialog>>add images Then again add image through same process
- 25. 24 | P a g e Click process in the bar and run mosaic>>select output folder Open raster layer>>add mosaic image In EarthExplorer maximum satellite images are geometrically corrected.So it is not need to do geometric correction in details.
- 26. 25 | P a g e CHAPTER SEVEN (IMAGE INTERPRETATION) Inquire Tools Home>>select>>inquire tools>>open image>>select feature in map Then evaluate the pixel value with band combination. In Landsat 5 ,false color composition 4,3,2 is a popular combination where red indicate vegetation ,blue indicates water.so in the cursor is in waterbody the pixel value of blue is larger.
- 27. 26 | P a g e Measurement Click to measure position, length, direction, area, and so forth, and optionally save the measurements to an annotation layer.Measurements can be based on a Cartesian system or on the ellipsoid. Home>>measure>>select parameter such as sq. Kilometer, meter etc.(a) select point>>polyline>>Set the line on a feature or objectives and measure(b) (a) (b)
- 28. 27 | P a g e Select point>>polygon>>draw a polygon of a feature and then measure.The measurement dialog show it. NDVI This is done in order to suppress, or normalize, varying effects such as viewing angles, sun shading, atmospheric effects, soil difference, and so on. It is also applied to maximize sensitivity to the feature of interest, such as the relative health of vegetation. To achieve this most Indices go beyond simple band division to include differencing, weighting, and the introduction of other variables Go to raster>>unsupervised>>NDVI
- 29. 28 | P a g e Select the sensor such as Landsat Tm and make sure that NIR and RED bands are selected Select input and output file Home>>inquire tools>>placed on this on ndvi image. It measure the file pixel and the range is +1>0>-1 . Less than zero means there is no vegetation
- 30. 29 | P a g e Profile Tools Spectral Profile Use the Spectral Profile dialog to visualize the reflectance spectrum of a single pixel through many bands. This technique is particularly useful for hyperspectral data that can have hundreds of layers. This technique allows estimates of the chemical composition of the material in the pixel. You can compare the profiles that you generate to those from laboratory (or field) spectrophotometers. Go to multispectral>>spectral profile Select the inquire tools and plot the features.It shows the pixel value for different layer.
- 31. 30 | P a g e Surface Profile Use the Surface Profile Viewer to visualize the reflectance spectrum of a rectangular area of data file values in a single band of data. You can overlay the wireframe surface with a grayscale, thematic, or true color image. Go to multispectral>>spectral>>surface profile Select feature and plot different layers.It shows 3d model of pixel values of different features.
- 32. 31 | P a g e Spatial Profile Use the Spatial Profile Viewer to visualize the reflectance spectrum of a polyline of data file values in a single band of data (one-dimensional mode) or in many bands (perspective three-dimensional mode). The most common example of single band data profile is that of a Digital Elevation Model (DEM) being used to create a height cross-section profile along a route. This helps in interpreting changes in elevation along a planned route and in identifying sections of the route which are particularly steep or flat. Go to multispectral>>surface profile>>spatial profile It shows pixel values with distance that you draw, in a plotted layer.
- 33. 32 | P a g e Unsupervised Classification ERDAS IMAGINE uses the ISODATA algorithm to perform an unsupervised classification. ISODATA stands for Iterative Self-Organizing Data Analysis Technique. It is iterative in that it repeatedly performs an entire classification (outputting a thematic raster layer) and recalculates statistics. Self-Organizing refers to the way in which it locates the clusters that are inherent in the data. Go to raster>>unsupervised>>unsupervised classification>>input and output files destination>>define classes in isodata>>define iteration, means scanning the main image Open raster layer and open that unsupervised image. Go to>>table>>show attribute>>edit class name,color (click on the icon) and also opacity by clicking on it
- 34. 33 | P a g e CHAPTER EIGHT (MAP PRODUCTION) Supervised Classification Supervised Classification provides tools for categorizing pixels using interactive supervised techniques. You provide examples of what particular classes look like, which are then used by the software algorithms to derive rules for mapping all other pixels into the class values.It is done by manually. Go to raster>>supervised>>signature editor>> Home>>drawing tools>>create polygon and add them as a class in signature features by clicking create new signature as AOI>>again do it of same feature but in another location. When many classes of one feature added, merge them(a) make another one class automatically ,delete the previous selection.(b) (A)
- 35. 34 | P a g e (b) Again create polygon of another feature by drawing tools and add them in signature tools as class like previous method. doing it one by one and add classes in signature tools. Edit the class name and color also by click it such as water-blue color After completing manually drawing method, save signature file as .sig. Then go to supervised classification>>locate input,output and .sig files.(a) After doing that open raster layer>>open the supervised image.(b) Go to home>>table>>open attribute table
- 36. 35 | P a g e (a) (b) It is done by manually when one can go to the location and analyses the features and observe them.Finally show the analyzed features on map by different classes but in unsupervised classification software done it automatically by stablished algorithm.
- 37. 36 | P a g e Unsupervised Classification See details in page no-32 There is a unsupervised classification with 5 classes such as water,bare land,vegetation,crop,settlement. Identify them with different colors in map . Add views>>reate new map view There add a viewer,click on map view,go to layout >>map frame Select map template and edit the map tile also
- 38. 37 | P a g e Double click on map viewer,after adding map frame,make sure that there is another viewer where unsupervised image is opened.Placed the unsupervised image on the box and click ok of the map frame tools. The unsupervised map is open on the map viewer
- 39. 38 | P a g e Layout>>map grid>>map grid Change margin>>length outside as your wish
- 40. 39 | P a g e Add the scale bar properties and select parameter such as meter. Select Legend option and identify the classes. Modify yourself
- 41. 40 | P a g e Add the north line Add locational map also Final map is ready File>>print THE END