Geometric Correction Preprocess RS data to remove geometric distortion so that individual picture elements pixels are in their proper planimetric x y map locations Geometrically corrected images for extraction of accurate distance polygon area and direction bearing of information Internal and External Geometric Error Scanning Cell Variation in Ground Resolution Size Systematic predictable or nonsystematic random Internal Geometric Errors introduced by RS system itself or in combination with Earth rotation or curvature characteristics Often systematic predictable and may be identified and corrected using pre launch or in flight platform ephemeris Skew caused by Earth rotation effects Scanning system induced variation in ground resolution cell size Scanning system one dimensional relief displacement And scanning system tangential scale distortion Orbital muiltspectral scanning system scans through just a few degrees off nadir as it collects data hundreds of kilometers above Earth s surface minimizes amount of distortion introduced by scanning system Suboribtal multispectral scanning may be operating just tens of kilometers AGL with a scan of FOV of perhaps 70 degrees introduced numerous types of geometric distortion that can be difficult to correct External Geometric Error usually introduced by phenomena that vary in nature through space and time Most important external variables that can cause geometric error in RS data are radom movements by aircraft or space craft at time of data collection Altitude changes and or Attitude changes roll pitch ya Ground Control Points Geometric distortions due to sensor system attitude and or altitude changes can be corrected using ground control points and mathematical models Ground Control point location on earth s surface like a road intersection that can be identified on image and located accurately on a map obtain two distinct sets of coordinates associated with each GCP Image coordinates specified in i rows and j columns and Map coordinates eg xy in feet in a state of plane coordinate system or meters in a UCM Paired coordinates ij and xy from many GCPs can be modeled to derive geometric transformation coefficients coefficients may be used to geometrically rectify RS image to standard datum and map projection Sources for GCP map coordinate information for image to map rectification include Hard copy planimetric maps GCP coordinates are extracted using ruler measurements or a coordinate digitizer Digital planimetric maps usgs digital topographic map series where GCP coordinates are extracted directly from a digital map on the screen Digital orthophotoquads that are already geometrically rectified eg USGS digital orthophoto quarter quadrangels and or GPS GNSS instruments may obtain coordinates of objects in the field Types of Geometric Correction Commerical RS data have much of systematic error removed Unless otherwise processed unsystematic random error may remain in image making it non planimateric pixels are not in correct xy planimetric map position Two geometric correction procedures are often used by scientists Image to map rectification and Image to image registration General rule of thumb rectify RS image to standard map projection Thuse well focus on image to map rectification Image to Map Rectification Process by which geometry is made planimetric For accurate area direction and distance measreuments image to map geometric rectification should be performed May not remove al distortion caused by topographic relief displacement Image to map rectification normally involves selecting GCP image pixel coordinates row and column with their map coordinate counterparts eg meters northing and easting in a UTM map projection Image to Image Registration Translation and rotation alignment process by which two images of like geometry and of same geographic area are positioned coincident with respect to one another so that corresponding elements of the same ground area appear in the same place on registered images Used when it is not necessary to have each pixel assigned a unique x y coordinate in map projexn eg to make cufrsory examinaxn of change involving two images obtained on different dates Hybrid approach to Image Rectification and Registration Same general principles are used in both image rectification and registration The difference in image map rectification reference is a map in a standard map projection whereas in image to image registration reference is another image If rectified image is used as reference base rather than traditional map any image registered to it will inherit geometric errors existing in reference image Most RS research used data rectified to map base For rigorous change detection between to or more image dates may be useful to select hybrid approach to involving both image to map rectification and image to image registratio Image to Map Geometric Recitification Logic Two base operations must be performed to geometrically rectify RS image to map coordinate system and Spatial interpolation Intensity interpolation Geometric relationship between input pixel coordinates c and row xy and asociatede map coordinates of this same point xy must be identified A number of GCP pairs are used to establish nature of geometric coordinate transformation that must be applied to rectify or fill every pixel in output image xy with a value from a pixel in unrectified input image x y Pixel brightness values must be determined No direct one to one relaitonship between movement of input pixel values to output pixel locations A pixel in rectified output image often requires a low value from input pixel grid that does not fall neatly on a row and column coordinate Thus must be mechanism for determining brightnes value BV to be assigned to output rectified pixel process is called intensity interpolation Spatial Interpolation using Coordinate Transformations Image to map rectification polynomial equations fit to GCP data using least square criteria to model corrections directly in image domain without explicitly identifying source of distortion Depending on image distortion number of GCPs and degree of topographic relief displacement higher order polynomial equations may be required to geometrically correct the data Order of rectification is the highest exponent used in the polynomial Concept of how different order transformations fit a hypothetical surface illustrated in cross section Original observations First order linear transformation