Registration workshop report
The state-of-the-art in registration and rectification of image data for terrestrial applications is examined and recommendations for further research in these areas are made.
Ramapriyan, H. K.↗
SEARCH · Engineering Papers
Results for “registration”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
A translational registration system for LANDSAT image segments
The use of satellite images obtained from various dates is essential for crop forecast systems. In order to make possible a multitemporal analysis, it is necessary that images belonging to each acquisition have pixel-wise correspondence. A system developed to obtain, register and record image segments from LANDSAT images in computer compatible tapes is described. The translational registration of the segments is performed by correlating image edges in different acquisitions. The system was constructed for the Burroughs B6800 computer in ALGOL language.
Parada, N. D. J.↗
Assessment of Thematic Mapper band-to-band registration by the block correlation method
Rectangular blocks of pixels from one band image were statistically correlated against blocks centered on identical pixels from a second band image. The block pairs were shifted in pixel increments both vertically and horizontally with respect to each other and the correlation coefficient to the maximum correlation was taken as the best estimate of registration error for each block pair. For the band combinations of the Arkansas scene studied, the misregistration of TM spectral bands within the noncooled focal plane lie well within the 0.2 pixel target specification. Misregistration between the middle IR bands is well within this specification also. The thermal IR band has an apparent misregistration with TM band 7 of approximately 3 pixels in each direction. The TM band 3 has a misregistration of approximately 0.2 pixel in the across-scan direction and 0.5 pixel in the along-scan direction, with both TM bands 5 and 7.
Card, D. H.↗
Data management procedures for Tiepoint Registration, pre and post processing, and ICD116
The data management procedures for tiepoint registration, pre and post processing, and "ICD116' are described. With each procedure description, the pertinent execs are listed and purposes defined. An example run of each of the 32 execs is included with user inputs identified.
Nowakowski, B. S.↗
Seasat/SIR-A digital registration over Algeria
Seasat Synthetic Aperture Radar (SAR) and Shuttle Imaging Radar (SIR-A) data over northeastern Algeria have been studied using digital registration. Preliminary results of this work show that: (1) both radars delineate in the same way the same features on the ground, (2) the Seasat image shows a strong difference between the Chott Melrhir and its environment, not seen in the same way by SIR-A, and (3) SIR-A data enhance more details on the floor of the Chotts, presumably due to the look angle or due to flooding conditions which may have occurred prior to the SIR-A experiment (November 12-14, 1981).
Rebillard, P.↗
LANDSAT-4 MSS interband registration
Interband registration on raw LANDSAT-4 MSS data was measured in the four MSS bands of the Mistassini, Quebec scene (path-row 16-24, June 21, 1983). Statistical comparison between line-pixel locations of uniformly distributed ground control points in the four different bands permitted quantitative measures of the offset over the entire scene. The statistical distribution of the offset measures permitted evaluation of the standard error on the mean values, giving confidence on their precision. Results obtained using a manual ground control point done on a digital image correction system and those obtained using digital band-to-band correlation adapted from a digital stereographic correlation algorithm are tabulated. The pixel misrepresentation values obtained are compared with published NASA figures.
Grieve, G.↗
LANDSAT-MSS multitemporal registration
The multitemporal registration of LANDSAT-4 MSS products was tested for two different geocoded subscenes acquired in 1982 and 1983. The scenes were then compared with LANDSAT-2 MSS geocoded data from 1981. Statistical comparison between line-pixel locations of uniformly distributed ground control points in the three geocoded products permitted quantitative measure of the offset over the sub-scene. Values of line and pixel misregistration obtained using the manual ground control points and digital correlation are tabulated and discussed.
Grieve, G.↗
Automatic digital image registration
This paper introduces a general procedure for automatic registration of two images which may have translational, rotational, and scaling differences. This procedure involves (1) segmentation of the images, (2) isolation of dominant objects from the images, (3) determination of corresponding objects in the two images, and (4) estimation of transformation parameters using the center of gravities of objects as control points. An example is given which uses this technique to register two images which have translational, rotational, and scaling differences.
Goshtasby, A.↗
Automatic terrain elevation mapping and registration
Optimum radar illumination geometries for stereoscopic analysis of surface topography are determined. Correlation and image processing experiments on synthetic aperture radar (SAR) data for improved information extraction are conducted. Model of the geometry of the multiple SIR-B views of the Earth are developed the sensitivity of the derived terrain altitude data to the various system parameters is established. The limits of accuracy of terrain data achievable with shuttle imaging radar (SIR-B) are derived. Algorithms for matching multiple SIR-B images to generate digital terrain maps are developed. Finally, the use of such terrain maps in geometric correction and registration of SIR-B and LANDSAT Thematic Mapper data is demonstrated.
Ramapriyan, H. K.↗
Assessment of Thematic Mapper Band-to-band Registration by the Block Correlation Method
Rectangular blocks of pixels from one band image were statistically correlated against blocks centered on identical pixels from a second band image. The block pairs were shifted in pixel increments both vertically and horizontally with respect to each other and the correlation coefficient to the maximum correlation was taken as the best estimate of registration error for each block pair. For the band combinations of the Arkansas scene studied, the misregistration of TM spectral bands within the noncooled focal plane lie well within the 0.2 pixel target specification. Misregistration between the middle IR bands is well within this specification also. The thermal IR band has an apparent misregistration with TM band 7 of approximately 3 pixels in each direction. The TM band 3 has a misregistration of approximately 0.2 pixel in the across-scan direction and 0.5 pixel in the along-scan direction, with both TM bands 5 and 7.
Card, D. H.↗
Modular detector for deep underwater registration of muons and muon groups
Registration and identification of muons and muon groups penetrating into the ocean depth, can be performed using a modular multilayer detector with high resolution bidimensional readout - deep underwater calorimeter (project NADIR). Laboratory testing of a prototype sensor cell with liquid scintillator in light-tight casing, testifies to the practicability of the full-scale experiment within reasonable expences.
Demianov, A. I.↗
Registration of rotated images by invariant moments
A new technique for registration of rotated images using invariant moments as the similarity measure is given. If circular windows with invariant moments are used, then window search can be carried out on rotated images in the same manner cross-correlation with rectangular windows is used on translated images. Normalization of invariant moments for more accurate window similarity measurement is also given. The proposed approach is tested against two images from different satellites that have translational and rotational differences.
Goshtasby, A.↗
Fracture analysis of radial scientific instrument module registration fittings of the space telescope
Various pieces of the registration fittings for the Radial Scientific instrument (SI) module of the Space Telescope were examined from a fracture mechanics point of view and deemed to be fail-safe or else have had maximum allowable flaw sizes specified for them. The results of these analyses are summarized in tabular form. In many instances the applied stress levels were so low that the threshold stress intensity factor range was never reached. In most of the others the allowable flaw sizes were large enough to be detected by visual inspection. However, for some parts, such as the flexures connecting the aluminum cover to the ball retainer in the fitting at point A, the flaw sizes were rather small. Eddy current tests are capable of detecting flaws of this size (0.022 inches x 0.1 inches), so for those which have been so tested these small flaws should represent no danger of going undetected. In every instance approximations were made to err on the conservative side. These were pointed out in the discussions of the analyses for each fitting. One conservative approximation that was not mentioned, however, is the fact that retardation was not included in the crack propagation computations.
Springfield, C. W., Jr.↗
Simulation of electronic registration of multispectral remote sensing images to 0.1 pixel accuracy
Band-to-band coregistration of multispectral remote sensing images can be achieved by electronic signal processing techniques rather than by costly and difficult mechanical alignment. This paper describes the results of a study of the end-to-end performance of electronic registration. The software simulation includes steps which model the performance of the geometric calibration process, the instrument image quality, detector performance and the effects of achieving coregistration through image resampling. The image resampling step emulates the Pipelined Resampling Processor, a real-time image resampler. The study demonstrates that the electronic alignment technique produces multispectral images which are superior to those produced by an imager whose pixel geometry is accurate to 0.1 pixel rms. The implications of this approach for future earth observation programs are discussed.
Reitsema, H. J.↗
Subpixel translation-registration of random fields
The author examines both theoretically and through a simulation study the feasibility of identifying the location within a large reference gray-level array of a smaller sensed array to an accuracy finer than one pixel. It is assumed that the sensed image before discretization into pixels consists of a translated, but not rotated, section of the reference image with some superposed noise. The theoretical and empirical results show that when the noise has standard deviation no larger than that of a realistic reference field, the upper quartile of the registration error is on the order of 0.25-0.5 pixels.
Slud, Eric V.↗
GOES I/M image navigation and registration
Image Navigation and Registration (INR) is the system that will be used on future Geostationary Operational Environmental Satellite (GOES) missions to locate and register radiometric imagery data. It consists of a semiclosed loop system with a ground-based segment that generates coefficients to perform image motion compensation (IMC). The IMC coefficients are uplinked to the satellite-based segment, where they are used to adjust the displacement of the imagery data due to movement of the imaging instrument line-of-sight. The flight dynamics aspects of the INR system is discussed in terms of the attitude and orbit determination, attitude pointing, and attitude and orbit control needed to perform INR. The modeling used in the determination of orbit and attitude is discussed, along with the method of on-orbit control used in the INR system, and various factors that affect stability. Also discussed are potential error sources inherent in the INR system and the operational methods of compensating for these errors.
Fiorello, J. L., Jr.↗
An accurate registration technique for distorted images
Accurate registration of International Ultraviolet Explorer (IUE) images is crucial because the variability of the geometrical distortions that are introduced by the SEC-Vidicon cameras ensures that raw science images are never perfectly aligned with the Intensity Transfer Functions (ITFs) (i.e., graded floodlamp exposures that are used to linearize and normalize the camera response). A technique for precisely registering IUE images which uses a cross correlation of the fixed pattern that exists in all raw IUE images is described.
Delapena, Michele↗
Simulation studies of the GOES-I Image Navigation and Registration (INR) system
The GOES-I Image Navigation and Registration (INR) system will fly on the next generation of NOAA geostationary meteorological satellites. This system uses observations of stars and earth landmarks made by the Imager and Sounder Instruments themselves to determine the orbit, attitude, and misalignment state of the spacecraft and instruments. An extensive program of simulation and analysis has been conducted to assess the expected performance of the INR system. This paper discusses these activities and presents some computer simulation results showing expected system performance under a variety of conditions including station-keeping.
Carr, James L.↗