Digital Image Processing Review
Digital Image Processing Review GPY 470
Popular in Digital Image Processing
Popular in Geography
This 7 page Study Guide was uploaded by Maddy Moldenhauer on Thursday February 25, 2016. The Study Guide belongs to GPY 470 at Grand Valley State University taught by Sun in Winter 2016. Since its upload, it has received 51 views. For similar materials see Digital Image Processing in Geography at Grand Valley State University.
Reviews for Digital Image Processing Review
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 02/25/16
Lecture Remote Sensor Properties 1 What are the four properties of remote sensors? Spectral, Spatial, Radiometric, and Temporal Resolution 2 What is spectral resolution? The ability of a sensor to differentiate between EM radiation of different wavelengths (expressed in terms of # of bands). The finer the spectral resolution, the narrower the wavelength range, the more spectral bands. 3 What is spatial resolution? The size of the smallest object that can be detected. Determined by pixel size (expressed in meters). The higher (finer) the resolution, the smaller the number. 4 What is radiometric resolution? Ability of a sensor to detect differences in brightness value levels (expressed in the # of bits). The higher the radiometric resolution, the larger the # of bits. 5 What is temporal resolution? How often a RS system images a given area (defined by the time interval between repeat coverage of a given area). Lecture questions Radiometric correction 1. What is radiometric correction? What are the three errors addressed in radiometric correction? Addresses variations (errors) in the measured brightness value of the pixels due to sensor errors, atmospheric effects, and topographic effects. 2. What are sensor errors? What are the two common types of sensor errors? Introduced when the individual detectors do not function properly, the errors are usually corrected/calibrated at the satellite ground stations. Common sensor errors are line dropouts and striping or banding. 3. What is line dropout? Occurs when a detector completely fails to function or becomes temporarily saturated during a scan. This results with a brightness value of 0 (black) or 255 (white) for every pixel in a particular line and a particular band. 4. What is striping or banding? Occurs if a detector goes out of adjustment, it provides readings consistently great than or less than the other detectors for the same band over the same ground cover. To correct the data, we can identify the bad scan lines by computing histograms for detectors, bias correction, highpass filtering, forward and reverse principal component transformations. 5. What are atmospheric effects? The impact of the atmosphere on the remotely sensed data is called atmospheric effects. All electromagnetic radiation, before reaching sensors, passes through some distance or path length of the atmosphere. This is caused by atmospheric scattering and absorption on electromagnetic radiation. This minimizes the ability of extracting useful terrain information and must be eliminated for scene matching and change detection analysis. 6. What are the 3 types of reflectors? Specular (mirrorlike) : angle of incidence = angle of reflection Lambertian (diffuse) : reflects energy uniformly in all directions Regular case: most bodies behavior between the ideal specular and diffuse reflector 7. What is a digital number? Is a digital number a physical quantity? Scaled integers (brightness values) from quantization, which is not a physical quantity. Ex. 8 bit quantization system has digital numbers that can be any integer in the range 0255. 8. What is radiance? Is it a physical quantity? What is the unit of radiance? The most precise remote sensing radiometric measurement, which is a PHYSICAL quantity. Physical quantity has a sort of measuremental unit, in this case radiance is measured in watts per meter squared per steradian. 9. What is sensor calibration? 10. Suppose Band 7 of Landsat 2 MSS sensor has the following coefficients in the calibration equation: Lλ = 0.603*DN + 1.1, where L is λ diance and DN is digital number. DN=100 L λ= 0.603*100 + 1.1= 61.4 (Wm sr )2 1 If DN = 100, calculate the radiance for band 7. 11. Is the solar irradiance the only energy source incident on the surface of interest? No 12. What are the two major components of global irradiance incident on the surface? Direct solar irradiance (Electromagnetic radiation) and Diffuse sky irradiance (iradiance comes from scattered sunlight/atmosphere) 13. Which two elements are included in diffuse sky irradiance? EM from the Sun scattered by the atmosphere before the EM reaches Earth’s surface EM from the surrounding atmosphere area 14. What are the two major components of radiance reaching at the sensor? Transmitted radiance from the target toward the sensor and path radiance (atmospheric path radiance and background radiance) . 15. Which two elements are included in path radiance? Atmospheric path radiance: radiation from the sun scattered b the atmosphere, Background radiance: radiation reflected by nearby targets 16. What are the two broad types of atmospheric correction? Absolute atmospheric radiometric correction Relative atmospheric radiometric correction 17. What is reflectance? What is the unit of reflectance? The ratio of the total amount of radiation, as of light, reflected by a surface to the total amount of radiation incident on the surface. Percentage % Questions Lecture Spectral Enhancement Questions Work in small groups of 2-3 students, discuss and answer the following questions: 1 What is spectral enhancement? Enhancing images by transforming the values of each pixel on a multiband basis. Can be done on a single multispectral image, or two or more images of the same area acquired at different times. 2 What is the other term for spectral enhancement? Image Transformation 3 What is image subtraction? Give examples. A common operation applied to image data that is used to identify changes that have occurred between images collected on different dates. 4 What are Vegetation Indices? A “synthetic image layer” created from the existing bands of a multispectral image. This VI image layer provides unique and valuable information not found in any of the other individual bands. For Example, VI’s have been shown to quantify or predict vegetation biomass, productivity, leaf area, and/or vegetative ground cover. 5 In Simple Ratio, which spectral bands are used? What is the equation? Landsat: Band 4/Band 3 6 What is NDVI (Normalized Difference Vegetation Index? Show the equation. Values ranging from -1 to 1 Water = Neg. values Vegetated Areas = higher values Rock/dry soil = values near zero 7 What is EVI (Enhanced Vegetation Index)? Has improved sensitivity to high biomass regions, C1 and C2 coefficients adjusting for atmospheric effects and L is a soil adjustment factor 8 What is principal component analysis (PCA)? A linear transformation of the original bands and can be used to reduce the dimensionality in the data. The “new” bands are called principle components that may be more interpretable than the original images. Lecture questions Digital change detection 1 What are the general steps for performing change detection? State the change detection problem (study area, frequency, classes) Considerations of important parameters when performing change detection Image processing of RS data to extract change information Access accuracy Distribute results 2 What are the remote sensing system considerations when performing change detection? The ROI (region of interest) must be carefully identified and held constant throughout the project, it must be completely covered by the dates of needed imagery, and avoid data voids Must have a long enough time period to capture the information of interest, choose optimal change detection time periods, careful selection of time periods Select an appropriate land cover/land use classification scheme RS Data must hold the following resolutions constant: 1. Temporal Resolution (same time of day to eliminate sun angle effects) 2. Spatial Resolution (Same = Ideal, different = must resample he lower res data to high) 3. Spectral Resolution (Ideal = the same sensor system used on multiple dates, band that approximate one another) 4. Radiometric Resolution (Same radiometric precision on both dates, 8bit, Different = lower res data is decompressed to higher res) 3 What are the environmental considerations when performing change detection? 1 Atmospheric conditions 2 Soil moisture conditions 3 Phonological cycle characteristics 4 Tide stage 4 Is change information extracted from raw remote sensing data or classified thematic map? You can do both! 5 What is the write function memory insertion method? Used to visually identify change in the imagery using additive color theory by placing images of different years in red, green and blue planes 6 What is image algebra? The use of images as variables in normal arithmetic operations (Band differentiating, band ratioing, etc). Identify the amount of change between two rectified images by image differencing (subtracting the imagery of one date from that of another). 7 What does each pixel value normally represent in a raw remote sensing image? The difference between the two rectified images brightness values 8 Given the following two raw remote sensing images from 2000 and 2010, find change between 2000 and 2010 using the image algebra change detection method. 0 5 12 11 0 0 8 1 1 11 13 0 5 0 0 8 2 80 12 0 5 8 60 13 12 0 0 3 9 20 18 5 0 0 0 8 24 108 105 10 9 95 112 1) image_2000 2) image_2010 20 60 180 210 8 10 2 10 2 9 2 25 2 11 15 13 1 30 20 30 5 9 What is postclassification comparison change detection? The difference between the two rectified images classes 10 What does each pixel value normally represent in a classified map/image? Feature Class 11 Given the following two land use/land cover (classified) maps from two different time periods, complete the table below, which is called change matrix. To: 2010 1 1 1 2 1 2 2 From: Class Forest (1) Water (2) Urban (3) 1 2 2 3 2000 Forest (1) 1 2 3 2 3 2 3 Water (2) 4 5 6 2 3 1 2 Urban (3) 7 8 9 2000 map 2010 map 2 2 3 2 1ege3d: 3 – F2rest, 2 – Water, 3 – Urban 1 1 2 22 3 C1ange2Image 2 2 6 8 4 3 6 8 2 3 1 2
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'