Remote sensing practice
Msc Shirinbek Mazambekov
University of West Hungary
Table of Contents
- Area explanation
- Downloading image
- Explanation of image
- Subset Image
- Statistical filter
- Focal Analysis
- Adaptive Filter
- Layer stack
- Principal Component
- Tasseled Cap
- Unsupervised Classification
- Image Segmentation
- Insert geometry
- Supervised Classification
- Signature editor
- Feature space image
- Feature space thematic
Tajikistan Kurgan-Tube district
Qurghonteppa (formerly known as Kurgan-Tyube) is an agricultural city on the Vaksh River in southwestern Tajikistan. It is the capital of the Viloyati Khatlon region and is located 100km from Dushanbe, with a population of 85,000 (2006), making it the third-largest city in the country. It lies along the main north-south route between Dushanbe and Afghanistan, on which a new bridge was built across the Pyanj River in 2006. The city is a major center for the production of cotton and other agricultural crops, having some of the flattest and most fertile land in the nation. The area is dominated by gently rolling hills, unusual patches of forest and endless cotton and wheat crops.
Total land area (ha) 26,965 Irrigated area (ha) 1607
Rainfed land (ha) 2.4 Arable land (ha) in 1607
Gardens (ha): 70.6 Vineyards (ha): 2.4
Mulberry (ha): 1.5 Meadows (ha): 0.0
Citrus (ha): 0.1 The area under the seedlings (ha): 0.4
Abandoned (ha): 0.5 Pastures (ha): 0.0
Agricultural land: 0.1 Buildings (ha): 8.7
Dehkan farms (kolkhoz kind): 15 Height above sea level (in measures) 450
ArcGISEarth representation of site
For satellite images, I used Landsat 8 free data EarthExplorer USGS
Dataset_Identifier: Landsat 8 OLI/TIRS
Band_Identification: Number_of_Bands: 8
Originator: U.S. Geological Survey (USGS) Earth Resources
Observation and Science (EROS)
Title: LANDSAT_8 - Path: 153 Row: 34 for Scene:
Geospatial_Data_Presentation_Form: Remote-Sensing Image
Publication_Place: Sioux Falls, South Dakota, USA
Publisher: U.S. Geological Survey (USGS) Earth Resources Observation and Science (EROS) Center
Creating subset image for all 7 bands
Subsetting refers to breaking out a portion of a large file into one or more smaller files. Often, image files contain areas much larger than a particular study area. In these cases, it is helpful to reduce the size of the image file to include only the area of interest (AOI). This not only eliminates the extraneous data in the file, but it speeds up processing due to the smaller amount of data to process. This can be important when dealing with multiband data.
Subset area for previewing or importing using map or Inquire Box for interesting area. And using Batch for all bands and rename Ignore Zero in output status I create separate 7 rasters.
Result of my 7 bands subsetting
Layer Stack The layer stack function is useful for placing layers one on top of the other. Also, Landsat 7files often come in one-band-per-image files, and this procedure will work as well, but keep in mind that Landsat 7 band 6 (thermal) typically has 60-meter resolution, and the panchromatic band typically has15-meter resolution, while the other bands have 30-meter resolution. If the panchromatic band is stacked into the output file, bands 1 through 7 will be resampled some spectral definition will be lost, and the file size more than quadrupled. For Landsat 7 data, layer stack bands 1through 5 and band 7. Keep the panchromatic band as separate files.
Convolution filtering is the process of averaging small sets of pixels across an image. Convolution filtering is used to change the spatial frequency characteristics of an image. A convolution kernel is a matrix of numbers that is used to average the value of each pixel with the values of surrounding pixels in a particular way. The numbers in the matrix serve to weight this average toward particular pixels. These numbers are often called coefficients because they are used as such in the mathematical equations. To compute the output value for this pixel, each value in the convolution kernel is multiplied by the image pixel value that corresponds to it. These products are summed, and the total is divided by the sum of the values in the kernel, as shown here:
Raster –Spatial—Convolution—Kernel 3*3 Edge detect
Raster –Spatial—Convolution—Kernel 5*5 Law pass
Statistical filter--produces the pixel output Digital Number by averaging pixels within a moving window that fall within a statistically defined range.
Focal analysis (SD)
Enables you to perform one of several analyses on class values in an image file using a process similar to convolution filtering.
Varies the contrast stretch for each pixel depending upon the DN values in the surrounding moving window.
Normalized Difference Vegetation Index (NDVI)=(NIR – R)/(NIR + R)
The normalized difference vegetation index (NDVI) is a simple graphical indicator that can be used to analyze remote sensing measurements, typically but not necessarily from a space platform, and assess whether the target being observed contains live green vegetation or not.
(NDVI) the formula for NDVI is IR-R/IR+R, where IR stands for the infrared portion of the electromagnetic spectrum and R stands for the red portion of the electromagnetic spectrum. NDVI finds areas of vegetation in imagery. [-1:1]
Tasseled Cap transformation
The Tasseled-Cap Transformation is a conversion of the original bands of an image into a new set of bands with defined interpretations that are useful for vegetation mapping. A tasseled-cap transform is performed by taking “linear combinations” of the original image bands - similar in concept to principal components analysis. So each tasseled-cap band is created by the sum of image band 1 times a constant plus image band 2 times a constant, etc… The coefficients used to create the tasseled-cap bands are derived statistically from images and empirical observations and are specific to each imaging sensor.
Tasseled-Cap Transformation —an image enhancement technique that optimizes data viewing for vegetation studies or rotates the data structure axes to optimize data viewing for vegetation studies.(I didn’t change the settings existing default )
Clay Minerals LS8TM
TM 6/TM7 area without vegetation
Clay minerals are hydrous aluminum phyllosilicates, sometimes with variable amounts of iron, magnesium, alkali metals, alkaline earths, and other captions found on or near some planetary surfaces. Clay minerals form in the presence of water and have been important to life, and many theories of abiogenesis involve them. They have been useful to humans since ancient times in agriculture and manufacturing.
Principal components analysis (PCA) is often used as a method of data compression. It allows redundant data to be compacted into fewer bands—that is, the dimensionality of the data is reduced. The bands of PCA data are non correlated and independent, and are often more interpretable than the source data(1 band) The process is easily explained graphically with an example of data in two bands. Below is an example of a two-band scatterplot, which shows the relationships of data file values in two bands. The values of one band are plotted against those of the other. If both bands have normal distributions, an ellipse shape results.
In supervised classification, it is important to have a set of desired classes in mind, and then create the appropriate signatures from the data. You must also have some way of recognizing pixels that represent the classes that you want to extract. Supervised classification is usually appropriate when you want to identify relatively few classes, when you have selected training sites that can be verified with ground truth data, or when you can identify distinct, homogeneous regions that represent each class.
Maximum Likelihood Pixels inside of a stated threshold (Standard Deviation Ellipsoid) are assigned the value of that class signature.Pixels outside of a stated threshold (Standard Deviation Ellipsoid) are assigned a value of zero (NULL) , Disadvantages: Much slower than the minimum distance or parallelepiped classification algorithms. The potential for a large number of NULL. Advantages: more “accurate” results (depending on the quality of ground truth, and whether or not the class has a normal distribution
Min distance Every pixel is assigned to the category based on its distance from cluster means. Standard Deviation is not taken into account. Disadvantages: generally produces poorer classification results than maximum likelihood classifiers. Advantages: Useful when a quick examination of a classification result is required.
The Mahalanobis distance is a measure of the distance between a point P and a distribution D, introduced by P. C. Mahalanobis in 1936.It is a multi-dimensional generalization of the idea of measuring how many standard deviations away P is from the mean of D. This distance is zero if P is at the mean of D, and grows as P moves away from the mean: along each principal component axis, it measures the number of standard deviations from P to the mean of D. If each of these axes is rescaled to have unit variance, then Mahalanobis distance corresponds to standard Euclidean distance in the transformed space. Mahalanobis distance is thus unit less and scale-invariant, and takes into account the correlations of the data set.
|Value||Class_Names||Count of pixel||Red||Green||Blue||Opacity||area ha|