An automated parallel image registration technique based on the correlation of wavelet features

Appl. Inf. Sci. Branch, NASA Goddard Space Flight Center, Greenbelt, MD
IEEE Transactions on Geoscience and Remote Sensing (Impact Factor: 2.93). 09/2002; DOI: 10.1109/TGRS.2002.802501
Source: IEEE Xplore

ABSTRACT With the increasing importance of multiple multiplatform remote sensing missions, fast and automatic integration of digital data from disparate sources has become critical to the success of these endeavors. Our work utilizes maxima of wavelet coefficients to form the basic features of a correlation-based automatic registration algorithm. Our wavelet-based registration algorithm is tested successfully with data from the National Oceanic and Atmospheric Administration (NOAA) Advanced Very High Resolution Radiometer (AVHRR) and the Landsat Thematic Mapper (TM), which differ by translation and/or rotation. By the choice of high-frequency wavelet features, this method is similar to an edge-based correlation method, but by exploiting the multiresolution nature of a wavelet decomposition, our method achieves higher computational speeds for comparable accuracies. This algorithm has been implemented on a single-instruction multiple-data (SIMD) massively parallel computer, the MasPar MP-2, as well as on the CrayT3D, the Cray T3E, and a Beowulf cluster of Pentium workstations.

1 Bookmark
  • Source
  • [Show abstract] [Hide abstract]
    ABSTRACT: Significant developments in the field of remote sensing have widened the application fields of remote sensing. One of the application fields is a land consolidation project. Multi-temporal and/or multi-sensor remote sensing provides a unique tool to track land utilization dynamics but requires precise registration of thousands of satellite images. However, automatic registration between remote sensing images is a challenging problem due to the different geometric distortion within the images, the illumination variation and varying resolution. To address this problem, we propose a hybrid automatic image registration scheme (technique), which combines Phase Correlation (PC) method and SIFT descriptor registration together. Based on the specific characteristic of the remote sensing imagery, we apply a Phase Correlation (PC) method first to coarsely pro-register the input image to the reference image. Then, a fine-scale registration process based on the scale invariant feature transform (SIFT) method is constructed. Experiments with Quickbird, CBERS-lremote-sensing images of the land Consolidation area in Beijing demonstrate that the proposed hybrid method is fully automatic and fast. Moreover, the registration accuracy is higher than traditional methods.
    Intelligent Automation and Soft Computing 01/2012; 18(8). DOI:10.1080/10798587.2008.10643316 · 0.19 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Investigation of urban built environment includes a wide range of applications that require 3-D information. New approaches are needed for near-real-time analysis of the urban environment with natural 3-D visualisation of extensive coverage. Hyperspectral remote sensing technology is a promising and powerful tool to assess quantitative classification of urban materials by exploring possible chemical/physical changes using spectral information across the VIS-NIR-SWIR spectral region. Light Detection And Ranging (LiDAR) technology offers precise information about the geometrical properties of the surfaces and can reflect the different shapes and formations in the complex urban environment. Generating a monitoring system that is based on integrative fusion of hyperspectral and LiDAR data may enlarge the application envelope of each individual technology and contribute valuable information on urban built environments and planning. A fusion process defined by a data-registration algorithm and including spectral/spatial and 3-D information is developed and presented. The proposed practical 3-D urban environment application for photogrammetric and urban planning purposes integrates the hyperspectral (spectrometer, ground camera and airborne sensor) and LiDAR data. This application may provide urban planners, civil engineers and decision-makers with tools to consider quantitative spectral information in the 3-D urban space.
    03/2013; 4(1). DOI:10.1080/19479832.2011.618469